September 3, 2012
Stat of the Week Competition: September 1 – 7 2012
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 September 7 2012.
- Statistics can be bad, exemplary or fascinating.
- The statistic must be in the NZ media during the period of September 1 – 7 2012 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.
The fine print:
- Judging will be conducted by the blog moderator in liaison with staff at the Department of Statistics, The University of Auckland.
- The judges’ decision will be final.
- The judges can decide not to award a prize if they do not believe a suitable statistic has been posted in the preceeding week.
- Only the first nomination of any individual example of a statistic used in the NZ media will qualify for the competition.
- Employees (other than student employees) of the Statistics department at the University of Auckland are not eligible to win.
- The person posting the winning entry will receive a $20 iTunes voucher.
- The blog moderator will contact the winner via their notified email address and advise the details of the $20 iTunes voucher to that same email address.
- The competition will commence Monday 8 August 2011 and continue until cancellation is notified on the blog.
Rachel Cunliffe is the co-director of CensusAtSchool and currently consults for the Department of Statistics. Her interests include statistical literacy, social media and blogging. See all posts by Rachel Cunliffe »
Statistic: Horror over child suicide rate surge. The number of suicides in the 15-19 age group has surged 40% in the latest year.
Source: New Zealand Herald
Date: 3 Sep 2012
My first reaction to this headline was that the “surge” was due to noise and not significant.
My second reaction was that there might be something real going on here. The difference between 56 and 80 is more than 3 standard deviations (80 – 56) / sqrt(56) = 3.2.
My third reaction after looking at the data (http://www.spinz.org.nz/file/downloads/pdf/file_529.pdf) is that it it just noise. For example there were 55 male suicides in the 15-19 age group and 25 females. The four previous years there were 40 and 16, 38 and 15, 35 and 25, 36 and 15.
Nothing unusual in the female numbers. Comparing the male figure of 55 with prior values of about 40 we have a jump of about 2 standard deviations. This isn’t statistically significant after we take into account that we cherry-picked the age group (out of 16) and the year (out of 5) to find this two standard deviation result.
We can’t really blame the Herald for this headline. The Coroner himself said “Significant is the jump in teenage suicide numbers” and he talked about trends which take into account more than just the two latest years and more than just one age group.
12 years ago
Statistic: Talks about the downward trend for Windows XP compared with Windows 8, especially since the later has taken over the former. What is interesting is their comment that 160 million users is a significant sample size… Is it? The samples were from one companies traffic – so could be skewed by the kind of websites they have and, for that matter, whether they are sampling the same 160 million?
[not to mention the issue of timelines and additional computers. E.g. does a person tend to upgrade their computer or, more typically, it becomes the old computer used for kids or a server and they add a new one with a new system.]
Mark
Source: Tech Report
Date: Sept 4, 2012
Challenging concepts of paired t-test?, timelines, sample size and [lack of] control group.
12 years ago
Statistic: Green Coffee Beans are a miracle weight loss cure
Source: TV3
Date: 4 Sep 2012
This is a good example of an examination of a clinical trial.
Dr Oz is a heart surgeon who has his own daytime TV show. The show is a strange mix of good medical advice and quackery.
This week he conducted a “clinical trial” where he gave a green coffee bean extract to 2 patients and monitored then for 5 days. Yup, that’s a big trial.
Both lost weight. As a result, a Google search for “green coffee bean extract” now returns 12 million hits. “Within minutes the Internet was buzzing with million of people asking where they could find this amazing product” reports squidoo.com.
One person lost 5 pounds in 5 days which is impossible to have been caused by the beans. This is 20,000 calories lost or 4000 calories per day. Since a normal human burns 2000 calories per day even fasting for 5 days would not produce that weight loss.
There is a single clinical trial that backs up the Dr Oz finding.
That trial is nicely debunked at http://www.sciencebasedmedicine.org/index.php/dr-oz-and-green-coffee-beans-more-weight-loss-pseudoscience/
This debunking is an informative and useful read (for stats students) which is the reason for this posting. The site itself is a useful resource for anyone wondering what works and what doesn’t.
12 years ago
Statistic: Should you wait months for sex?
Source: Stuff.co.nz
Date: 7 Sep 2012
This story claims that the longer you wait to have sex the better the sex is.
But a moment’s thought shows that the longer you wait for sex the longer the relationship has lasted and thus the better the quality of the relationship and the better the sex.
12 years ago