Pie chart of the week
Originally from jobvine.co.za, via a Harvard Business Review piece on data visualisation, this graph is supposed to show salary ranges for different positions.
It really doesn’t. Read the HBR story for a better version.
Originally from jobvine.co.za, via a Harvard Business Review piece on data visualisation, this graph is supposed to show salary ranges for different positions.
It really doesn’t. Read the HBR story for a better version.
Privacy concerns are leading to “virtual identity suicide” with large numbers of Facebook users deleting their accounts, according to new scientific research.
A study investigating the phenomenon identified privacy as the biggest reason people are turning against the social network giant.
The new scientific research (not linked, journal not named)
The primary source for our convenience sample of Facebook quitters was the Website of the online initiative Quit Facebook Day. On this Website, Facebook users had the possibility to announce their intention to delete their account on May 31, 2010, which was declared as the Quit Facebook Day.
And what was the point of Quit Facebook Day?
In our view, Facebook doesn’t do a good job in either department. Facebook gives you choices about how to manage your data, but they aren’t fair choices, and while the onus is on the individual to manage these choices, Facebook makes it damn difficult for the average user to understand or manage this. We also don’t think Facebook has much respect for you or your data, especially in the context of the future.
So, how surprising is it that Quit Facebook Day quitters are more concerned about privacy than people who keep using Facebook?
From One News NZ, a story about Bobby Wilcox, the team’s performance analyst, who has a PhD in Statistics from our department
She’s been one of the Silver Ferns most integral members for nine years, yet she’s largely anonymous outside…
[the video comes with a very annoying ad, sadly]
Today is the 120th anniversary of women’s suffrage in New Zealand, with commemorations in a range of places, including the Centenary fountain in Khartoum Place, Auckland
The petition for women’s suffrage, signed by about 24000 women, was submitted to Parliament in July 1893
and the names of the petitioners have been digitised and made available at New Zealand History Online.
I haven’t been able to work out exactly what the adult female population of NZ was at the time, but the digital yearbook says that there were 305287 non-Maori females, that 30.94% were married and 4.11% widowed, and that there were 67000 never-married females 15 and older. Depending on how many of the never-married 15+ were 21 or older, this gives perhaps 150000, so about 16% of the non-Maori adult female population signed the petition. That compares to modern petitions with about 2.4% of voting-age people opposing marriage equality and about 10% for the anti-asset-sales petition (though these are targeting the entire NZ voting population, not just women).
Presumably, rather more than 16% of women were in favour of getting the right to vote, but it’s always difficult to track people down and get them all to sign. In 1893 there wasn’t an alternative: sampling hadn’t been invented, and would likely have been impractical — certainly, calling random phone numbers wouldn’t have got you very far.
Today, we have much more accurate ways of estimating the proportion of people who support some government action. Petitions, like demonstrations, are mostly useful for signalling to the government that some issue they weren’t aware of is actually important. For example, the petition against animal testing for legal highs would have been effective to the extent that the government wasn’t aware people cared about the issue. For anyone who was aware this was a political issue, a well-conducted opinion poll would be more informative and should be both less expensive and more effective than a petition.
Referendum petitions, as in New Zealand and some parts of the US, are an example of this principle: if an issue can get the support of 10% of the NZ voting population, it’s probably important enough to be worth serious consideration and debate. The threshold is weaker in many places. For example, in California a petition need only get 5% of the number of people who voted in the last election for state governor, which currently comes to under 2% of the adult population.
Here are the team ratings prior to Round 6, 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.
Current Rating | Rating at Season Start | Difference | |
---|---|---|---|
Canterbury | 22.22 | 23.14 | -0.90 |
Wellington | 10.85 | 6.93 | 3.90 |
Auckland | 7.52 | 9.02 | -1.50 |
Counties Manukau | 3.95 | 4.36 | -0.40 |
Waikato | 2.60 | 5.25 | -2.70 |
Tasman | -1.16 | -6.29 | 5.10 |
Bay of Plenty | -2.98 | -1.96 | -1.00 |
Taranaki | -3.09 | 3.92 | -7.00 |
Otago | -3.59 | -4.44 | 0.80 |
Hawke’s Bay | -5.00 | -6.72 | 1.70 |
North Harbour | -6.52 | -7.43 | 0.90 |
Northland | -7.85 | -8.26 | 0.40 |
Manawatu | -9.45 | -8.97 | -0.50 |
Southland | -10.77 | -11.86 | 1.10 |
So far there have been 38 matches played, 30 of which were correctly predicted, a success rate of 78.9%.
Here are the predictions for last week’s games.
Game | Date | Score | Prediction | Correct | |
---|---|---|---|---|---|
1 | Taranaki vs. Hawke’s Bay | Sep 11 | 23 – 10 | 7.80 | TRUE |
2 | Wellington vs. Bay of Plenty | Sep 12 | 33 – 16 | 18.60 | TRUE |
3 | Manawatu vs. Southland | Sep 13 | 27 – 17 | 5.00 | TRUE |
4 | North Harbour vs. Tasman | Sep 13 | 23 – 12 | -3.10 | FALSE |
5 | Waikato vs. Auckland | Sep 14 | 42 – 24 | -3.90 | FALSE |
6 | Canterbury vs. Otago | Sep 14 | 32 – 22 | 34.20 | TRUE |
7 | Counties Manukau vs. Taranaki | Sep 15 | 44 – 7 | 6.70 | TRUE |
8 | Hawke’s Bay vs. Northland | Sep 15 | 31 – 26 | 7.80 | TRUE |
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.
Game | Date | Winner | Prediction | |
---|---|---|---|---|
1 | Bay of Plenty vs. Southland | Sep 18 | Bay of Plenty | 12.30 |
2 | Auckland vs. Northland | Sep 19 | Auckland | 19.90 |
3 | Otago vs. Manawatu | Sep 20 | Otago | 10.40 |
4 | Counties Manukau vs. Waikato | Sep 21 | Counties Manukau | 5.90 |
5 | Wellington vs. Canterbury | Sep 21 | Canterbury | -6.90 |
6 | Taranaki vs. Bay of Plenty | Sep 21 | Taranaki | 4.40 |
7 | Tasman vs. Hawke’s Bay | Sep 22 | Tasman | 8.30 |
8 | Southland vs. North Harbour | Sep 22 | Southland | 0.20 |
Here are the team ratings prior to Round 7, 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.
Current Rating | Rating at Season Start | Difference | |
---|---|---|---|
Western Province | 4.17 | 4.47 | -0.30 |
Lions | 2.52 | -1.22 | 3.70 |
Sharks | 2.51 | 3.24 | -0.70 |
Cheetahs | -1.73 | -2.74 | 1.00 |
Blue Bulls | -3.89 | 0.59 | -4.50 |
Griquas | -5.73 | -6.48 | 0.80 |
So far there have been 18 matches played, 9 of which were correctly predicted, a success rate of 50%.
Here are the predictions for last week’s games.
Game | Date | Score | Prediction | Correct | |
---|---|---|---|---|---|
1 | Cheetahs vs. Lions | Sep 13 | 26 – 23 | 3.30 | TRUE |
2 | Griquas vs. Sharks | Sep 14 | 24 – 25 | -0.70 | TRUE |
3 | Blue Bulls vs. Western Province | Sep 14 | 18 – 29 | 1.70 | FALSE |
Here are the predictions for Round 7. 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.
Game | Date | Winner | Prediction | |
---|---|---|---|---|
1 | Western Province vs. Griquas | Sep 20 | Western Province | 17.40 |
2 | Sharks vs. Cheetahs | Sep 21 | Sharks | 11.70 |
3 | Lions vs. Blue Bulls | Sep 21 | Lions | 13.90 |
Here are the team ratings prior to this week’s games, 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.
Current Rating | Rating at Season Start | Difference | |
---|---|---|---|
Roosters | 11.32 | -5.68 | 17.00 |
Storm | 9.31 | 9.73 | -0.40 |
Sea Eagles | 7.70 | 4.78 | 2.90 |
Rabbitohs | 7.05 | 5.23 | 1.80 |
Cowboys | 6.60 | 7.05 | -0.40 |
Knights | 6.36 | 0.44 | 5.90 |
Bulldogs | 2.16 | 7.33 | -5.20 |
Titans | 2.11 | -1.85 | 4.00 |
Sharks | 1.89 | -1.78 | 3.70 |
Warriors | -0.89 | -10.01 | 9.10 |
Panthers | -2.57 | -6.58 | 4.00 |
Broncos | -5.15 | -1.55 | -3.60 |
Dragons | -8.18 | -0.33 | -7.90 |
Raiders | -10.23 | 2.03 | -12.30 |
Wests Tigers | -11.37 | -3.71 | -7.70 |
Eels | -19.87 | -8.82 | -11.00 |
So far there have been 196 matches played, 118 of which were correctly predicted, a success rate of 60.2%.
Here are the predictions for last week’s games.
Game | Date | Score | Prediction | Correct | |
---|---|---|---|---|---|
1 | Rabbitohs vs. Storm | Sep 13 | 20 – 10 | 0.30 | TRUE |
2 | Sharks vs. Cowboys | Sep 14 | 20 – 18 | -0.77 | FALSE |
3 | Roosters vs. Sea Eagles | Sep 14 | 4 – 0 | 3.52 | TRUE |
4 | Bulldogs vs. Knights | Sep 15 | 6 – 22 | 4.36 | FALSE |
Here are the predictions for Finals Week 2. 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.
Game | Date | Winner | Prediction | |
---|---|---|---|---|
1 | Sea Eagles vs. Sharks | Sep 20 | Sea Eagles | 5.80 |
2 | Storm vs. Knights | Sep 21 | Storm | 7.40 |
The Herald has a story about hazards of coffee. The picture caption says
Men who drink more than four cups a day are 56 per cent more likely to die.
which is obviously not true: deaths, as we’ve observed before, are fixed at one per customer. The story says
It’s not that people are dying at a rapid rate. But men who drink more than four cups a day are 56 per cent more likely to die and women have double the chance compared with moderate drinkers, according to the The University of Queensland and the University of South Carolina study.
What the study actually reported was rates of death: over an average of 17 years, men who drink more than four cups a day died at about a 21% higher rate, with little evidence of any difference in men. After they considered only men and women under 55 (which they don’t say was something they had planned to do), and attempted to control for a whole bunch of other factors, the rate increase went to 56% for men, but with a huge amount of uncertainty. Here are their graphs showing the estimate and uncertainty for people under 55 (top panel) and over 55 (bottom panel)
There’s no suggestion of an increase in people over 55, and a lot of uncertainty in people under 55 about how death rates differed by coffee consumption.
In this sort of situation you should ask what else is already known. This can’t have been the first study to look at death rates for different levels of coffee consumption. Looking at the PubMed research database, one of the first hits is a recent meta-analysis that puts together all the results they could find on this topic. They report
This meta-analysis provides quantitative evidence that coffee intake is inversely related to all cause and, probably, CVD mortality.
That is, averaging across all 23 studies, death rates were lower in people who drank more coffee, both men and women. It’s just possible that there’s an adverse effect only at very high doses, but the new study isn’t very convincing, because even at lower doses it doesn’t show the decrease in risk that the accumulated data show.
So. The new coffee study has lots of uncertainty. We don’t know how many other ways they tried to chop up the data before they split it at age 55 — because they don’t say. Neither their article nor the press release gave any real information about past research, which turns out to disagree fairly strongly.
From Pharmac’s annual report for 2012 (via @sudhvir), a graph comparing actual government expenditure on subsidised drugs (red) with what would be projected under pre-Pharmac subsidy policies (blue)
This table from the report shows some of how this was done. It shows the twenty drugs on which the most money was spent
Many of these are new (any drug whose name ends in ‘b’ is likely to be new), but they are mostly drugs that are genuinely better, at least for a subset of patients, than the alternatives. The top of the list, atorvastatin, lowers cholesterol more effectively than the cheaper simvastatin. It’s the best selling drug of all time, but in New Zealand is used only in a relatively small set of people whose cholesterol doesn’t go down enough on simvastatin. Adalimumab was a breakthrough in serious rheumatoid arthritis, and trastuzumab is the revolutionary breast cancer treatment sold as Herceptin.
Further down the list, candesartan is a blood pressure drug that can be used in people who have side-effects with some other blood pressure drugs. In Australia, candesartan and its relatives are used very widely; here they are used only when other drugs are insufficient or not tolerated.
Pharmac isn’t perfect, and I think it’s underfunded, but it does a very good job of getting most of the benefit of modern pharmaceutical medicine at a very low price.