Posts tagged Winner (58)

Thanks for your nominations last week in our Stat of the Week competition. We’ve selected Daniel Croft’s nomination of fascinating text message statistics on the advent of 20 years of text messaging:

“Wednesday was the 20 year anniversary of the first text message to have been sent, and so in response both Telecom and Vodafone have released some surprising information.

“Vodafone users sent 7.3 billion texts last year while Telecom was just under 7 billion. Thats 14,000,000,000 plus texts a year. Thats 38,000,000 a day spread between a population of 4,400,000… 8 texts per person a day. Or 2920 texts per person a year.

“A statistic that I would love to hear on this would be just how many of those 14 billion texts are just ‘LOL’.”

(A reminder that in our competition fine print, we state that we will not award Stat of the Week for a statistic coming from anyone at the University of Auckland outside the Statistics department. You can still nominate and discuss them, but the nomination won’t be eligible for the prize.)

Congratulations to Eva Laurenson for her excellent nomination of the NZ Herald’s article entitled “Manukau ‘luckiest’ place for Lotto”:

What does ‘luckiest’ in this title mean? Well to the average person ( I asked a few) they interpreted that title as ” I would have a higher chance of winning Lotto if I bought my ticket from a Manukau store compared to another store from a different suburb in Auckland.” Is this really the case? I doubt it. The article ranks Manukau ‘luckiest’ because it is the suburb with the highest total paid out first division amount. However no where did they take into account the total sales of Lotto tickets in each suburb. I think if you took this into account you’d see that Manukau sells alot more tickets than some of these other suburbs in Auckland. So even though Manukau can boast 55 mil in first division prizes we have no idea whether that is 55 mil out of 100 mill worth of ticket sales or 55 mil out of 1 bill worth of ticket sales. Some of the other suburbs may have a lesser amount of first division payouts compared to Manuaku but could have a greater proportion of first division payouts compared to ticket sales. Hence if that was true, your chance of winning first division given that you bought your ticket in that other suburb would be greater than (the same probability measured for) Manukau. Therefore I think there isn’t sufficient information provided to make this claim.

What I think the article could say is ‘given I won first division, the chances that I bought my ticket in Manukau are ____ times the chance that I bought it somewhere else.’ Something to this effect could be derived from the information presented by the herald article and it makes a bit of sense. Is this what the article wrote though? Not at all. They summarised this finding into “Manukau is the luckiest Lotto suburb in Auckland.” Please! This screams misleading. As discussed above, there simply isn’t enough information to justify labelling Manukau the ‘luckiest’ suburb for Lotto. People have a clear idea of what it means to be lucky and that generally is that they have an increased chance of winning. This is not the conclusion you can draw from the information they provided and in this case I believe the herald got it wrong.

I also think, although probably not the authors intentions, labelling Manukau as the ‘luckiest’ suburb has the danger of enticing people to spend more on Lotto. This article published earlier in the year by the NZ herald noted that “Many South Auckland suburbs featured among those which gambled away the most money. Mangere Bridge, Flat Bush, Manukau and Manurewa were in the top dozen suburbs.”
Even though the article was talking about the pokies, Lotto is just another form of gambling. We shouldn’t be condemming one and sending a rosy message about another, especially to communities who are struggling as it is.

Overall I think this should be the Stat of the week because using ‘lucky’ was a nice little pun but in effect mislead people regarding their chances of winning first division depending on where they bought their ticket.

Secondly it seems wrong to label a suburb ‘luckiest’ and potentially encourage a community to spend more on Lotto there when it is known that it is a compartively poorer area than other Auckland suburbs and spends alot of money on gambling as it is.

Thomas expanded on this, saying:

This looks as if it’s claiming that tickets bought in Manukau have been more likely to win. If this was true, it would still be useless, because future lotto draws are independent of past ones.

It’s even more useless because there is no denominator: not tickets sold, not people in the suburb, not even number of Lotto outlets in the suburb.

What the statistic, and the accompanying infographic, really identifies is the suburbs that lose the most money on Lotto. That’s why Manukau and Otara are ‘lucky’ and Mt Eden and Remuera are ‘unlucky’, the sort of willfully perverse misrepresentation of the role of chance that you more usually see in right-wing US outlets.

Congratulations to Andrew Mardon for being this week’s Stat of the Week winner for his nomination of:

“The higher a country’s chocolate consumption, the more Nobel laureates it spawns per capita, according to findings released in the New England Journal of Medicine.”

While the article mentions both correlation and causation, it doesn’t clearly explain the difference between the two and makes the distinction murkier by saying science is fallible and including a gem quote such as this:

“[The US] would have to up its cocoa intake by a whopping 125 million kg a year to produce one more laureate, said Franz Messerli, who did the analysis.”

Thanks for your nominations last week in our Stat of the Week competition. We’ve selected Alan Keegan’s nomination of car theft data which doesn’t take into consideration the number of vehicles when considering the likelihood of them being stolen.

(Thanks Brent for noticing the date typo in the competition, which has now been corrected. We’d like to think that Stats Chat is not about pointing out others’ typos or mistakes, but intelligently discussing the statistics behind the news and the issues surrounding them.)

Congratulations to Nick Iversen for her fantastic nomination of a “clinical trial” conducted by Dr. Oz with a great link to Science Based Medicine debunking the study.

James Curran noted that “it’s easy to loose 5lb in water even in the course of a day (on some bike rides in summer I will lose as much as 3-4kg on a 3 hour ride, and usually 1-2kg)”

Congratulations to Mark Bellhouse for his nomination of good reporting of a drug use study. Thomas Lumley also posted about the story here.

Congratulations to Eric Crampton for his nomination of an absolute shocker of a statistic this week. The printed story completely mangled the information in the press release. It even went further and attributed malicious intent to the researchers.

“67 Maori children died avoidable deaths every year, costing taxpayers $200 million annually.”

The original study says that the social costs of health disparities ranges from $62m-$200m, including measures of the value of statistical lives lost among Maori children. In no way is the larger figure close to a “Cost to the taxpayer” except in the odd sense that costs borne by the parents who cared about those children are included in the value of statistical lives lost and some of those parents may have been taxpayers.

Another case of “Economic Impact” or “Social cost” turning into “cost to the taxpayer” when handled by journalists.

Discussed at length:

http://www.tvhe.co.nz/2012/08/06/journalist-ideological-cant-read/

http://www.offsettingbehaviour.blogspot.co.nz/2012/08/i-hate-economic-impact-numbers.html

No winner this week but thank you for your nominations. Please continue to nominate!

Congratulations to Tony Cooper for his nomination of “Men over 50 nation’s biggest drinkers”, which Thomas Lumley picked up on and posted about as a result.

It’s a great example of how the media can get it so wrong – and in doing so has created a myth. Congratulations Tony!

Thanks to those who nominated in last week’s Stat of the Week competition.

This week we’re awarding Lindsay Mitchell’s nomination of the NZ Herald’s editorial which stated that “The vast majority of child abuse is perpetrated by men.”

There are numerous official sources that disprove this claim. For instance: Fig 7. Number of female and male abusers of specific types of abuse and neglect 2006 (pg 117) and Apprehensions for assault on a child (pg 10)

This isn’t a new bad stat, but keeps cropping up in the media.