Stats Chat

Q; Did you see yoghurt prevents cancer?

A: Where?

Q: The Herald (from the Daily Telegraph): “8 ways to lower your cancer risk.” Number one is “Eat yoghurt”. And they even have a link to research. How’s that for impressive?

A: Not exactly a link. They mention the name of a journal, but don’t even give the researchers’ names.

Q: Can’t you find them?

A: Of course. It’s even open-access.

Q: So, how much yoghurt did the people have to eat?

A: No yoghurt was harmed in this experiment. Also no people.

Q: Mice?

A: Mice.

Q: But yoghurt?

A: No. Some of the mice were set up with a restricted set of gut bacteria (missing known nasty ones) by being raised in a mouse colony who all had the restricted set.

Q: But the story says “gave one group of mice beneficial bacteria through probiotic supplements and the other non-beneficial bacteria. ”

A: Yes, it does. The research paper, not so much. Nor even the press release.

Q: So why yoghurt?

A: One of the bacteria that was more common in the mice with the restricted set is a Lactobacillus strain. Other Lactobacillus strains, even sometimes from the same species, are involved in making yoghurt, sourdough, sauerkraut, kimchi, etc.

Q: And you could use the mouse bacteria to make these foods?

A: In principle, probably, though you might not want to advertise it that way.

Q: So, the mice with more Lactobacillus were less likely to get cancer?

A: These were mutant mice who all get cancer, so that’s not really the question. They took longer to get cancer.

Q: So we can’t really be confident yoghurt would prevent normal mice from getting cancer?

A: No, it’s too soon to tell.

Q: Good thing normal mice don’t read the newspapers, then.

Team Ratings for Round 12

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 85 matches played, 58 of which were correctly predicted, a success rate of 68.2%.
Here are the predictions for last week’s games.

Predictions for Round 12

Here are the predictions for Round 12. 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.

The first data under the new real-estate ownership reporting scheme is out. The Herald has a story and also includes a full copy of the report.

So, what proportion of Auckland property sales were reported as being to China?

In Auckland, the level of foreign investment was slightly higher than the national level, at 4 per cent, or 474 properties. Nearly 60 per cent of these properties went to Chinese tax residents.

That’s 60% of 4%, or a bit under 2.5%.  Auckland is different; in the rest of New Zealand the majority of foreign (tax-residency) investors are Australians.

The LINZ report does a good job explaining the real limitations of `tax residence’ as a criterion, but it’s a lot better than any previous data we’ve had.

There were also questions about actual residency and intention to occupy a home, but these were harder to interpret because of property bought by companies or trusts, where the questions didn’t have a good answer.

I’d suggest starting with the report rather than the news coverage.

“Coffee today is like God in the Old Testament”, says John Oliver, reviewing the positive and negative headlines over the past year or so.  It’s excellent, if a little overblown in places.

On a related note, another site has fallen for the ‘cheese addiction/casomorphin’ hoax that we’ve seen before a few times. This time it’s Pharmacy Times.

Each week, we would like to invite readers of Stats Chat to submit nominations for our Stat of the Week competition and be in with the chance to win an iTunes voucher.

Here’s how it works:

• Anyone may add a comment on this post to nominate their Stat of the Week candidate before midday Friday May 13 2016.
• Statistics can be bad, exemplary or fascinating.
• The statistic must be in the NZ media during the period of May 7 – 13 2016 inclusive.
• Quote the statistic, when and where it was published and tell us why it should be our Stat of the Week.

Next Monday at midday we’ll announce the winner of this week’s Stat of the Week competition, and start a new one.

(more…)

The Herald has a headline “Emma and Noah continue to be tops for baby names”, with this link from the web front page

In fact, Noah was number 11 as a baby boy’s name, and Emma didn’t make the top hundred names for baby girls in New Zealand.  The top names in NZ, as in this Stuff story from the first week of January, were Oliver and Olivia. That story also had tables and graphs from the Dept of Internal Affairs data.

The new Herald story is about the USA, where they take longer to accumulate and release the baby-name data, but where they have the indefatigable Laura Wattenberg to make sure it gets publicised.

In fact, it’s kind of surprising how much difference there is between the US and NZ lists. Enough to make it worth pointing out in the story.  UK data won’t be out for another few months. Based on last year, it’s a bit more similar to NZ. Maybe we’ll get another story then.

Q: Did you see in the Herald that texting doesn’t help relationships?

A: That’s what they said, yes.

Q: And is it what they found?

A: Hard to tell. There aren’t any real descriptions of the results

Q: What did they do?

A: Well, a couple of years ago, the researcher had a theory that “sending just one affectionate text message a day to your partner could significantly improve your relationship.”

Q: So the research changed her mind?

A: Sounds like.

Q: That’s pretty impressive, isn’t it?

A: Yes, though it doesn’t necessary mean it should change our mind.

Q: It sounds like a good study, though. Enrol some people and regularly remind half of them to send affectionate text messages.

A: Not what they did

Q: They enrolled mice?

A: I don’t think there are good animal models for assessing affectionate text messages. Selfies, maybe.

Q: Ok, so that publicity item about the research is headlined “Could a text a day keep divorce away?”

A: Yes.

Q: Did they people about their text-messaging behaviour and then wait to see who got divorced?

A: It doesn’t look like it.

Q: What did they do?

A: It’s not really clear: there are no details in the Herald story or in the Daily Mail story they took it from.  But they were recruiting people for an online survey back in 2014.

Q: A bogus poll?

A: Well, if you want to put it that way, yes. It’s not as bogus when you’re trying to find out if two things are related rather than how common one thing is.

Q: <dubiously> Ok . And then what?

A: It sounds like they interviewed some of the people, and maybe asked them about the quality of their relationships. And that people who didn’t see their partners or who didn’t get affection in person weren’t as happy even if they got a lot of texts.

Q: Isn’t that what you’d expect anyway? I mean, even if the texts made a huge difference, you’d still wish that you had more time together or that s/he didn’t stop being affectionate when they got off the phone.

A: Pretty much. The research might have considered that, but we can’t tell from the news story. There doesn’t even seem to be an updated press release, let alone any sort of publication.

Q: So people shouldn’t read this story and suddenly stop any social media contact with their sweetheart?

A: No. That was last week’s story.

As you know, Leicester City won the English Premier League this week. At the start of the season, you could get 5000:1 odds on this happening. Twelve people did.

Now, most weeks someone wins NZ Lotto first division, which pays more than 5000:1 for a winning ticket, and where we know the odds are actually unfavourable to the punter. The 5000:1 odds on their own aren’t enough to conclude the bookies had it wrong.  Lotto is different because we have good reasons to know that the probabilities are very small, based on how the numbers are drawn. With soccer, we’re relying on much weaker evidence.

Here’s Tim Gowers explaining why 5000:1 should have been obviously too extreme

The argument that we know how things work from following the game for years or even decades is convincing if all you want to prove is that it is very unlikely that a team like Leicester will win. But here we want to prove that the odds are not just low, but one-in-five-thousand low.

Professor Gowers does leave half the question unexamined, though

I’m ignoring here the well-known question of whether it is sensible to take unlikely bets just because your expected gain is positive. I’m just wondering whether the expected gain was positive.

• Interactive exploration of colours used to label subway lines around the world

• A historical list of data visualisations, starting in 5500BCE Mesopotamia

• An example of rainbow colour scales actually misleading people

• Selection bias: @_OneRandomTweet retweets one random tweet every few hours, “to remind you that people don’t use Twitter like you do“

• New Zealand used to have open data on road crashes. Richard Law writes about why getting rid of it was bad.

• Richard Clark: “The boundaries of New Zealand suburbs and localities is held by the New Zealand Fire Service. For years, the NZFS has refused to provide this data under any terms except a restrictive license, and it has to stop.“

• Outsourcing: Andrew Gelman is disappointed in the NZ Herald, so I don’t have to be.

• Survivor bias:

pic.twitter.com/3LS7yH8adC

— Persian Rose (@PersianRose1) April 24, 2016

Team Ratings for Round 11

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 78 matches played, 53 of which were correctly predicted, a success rate of 67.9%.
Here are the predictions for last week’s games.

Predictions for Round 11

Here are the predictions for Round 11. 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.