123 meme

I've been tagged by This Humanist for a new blog meme. Here are the rules:

1. Pick up the book nearest you with at least 123 pages. (No cheating!)
2. Turn to page 123.
3. Count the first five sentences.
4. Post the next three sentences.
5. Tag five other bloggers.

Before me I have a draft copy of a superb statistics textbook for linguists. It's been giving me solace and relief as I struggle to extract meaning from a particularly messy set of measurements. The sentences in question come near the start of a section on interpreting statistical significance:

The random noise is obtained by adding to each y value, a random number from a normal distribution with mean 0 and a standard deviation of 80.

[I've left out the example code to execute in statistics software - see the book if you're interested.]

A simulation run will typically produce non-significant results, such as

[More code and output illustrating that the linear relationship is statistically invisible behind all the random noise.]

Although there is a linear relation between y and x - we built it into the data set ourselves - the amount of noise that we superimposed is so large that we cannot detect it.

The author then goes on to demonstrate that, by increasing the number of samples, we can increase the sensitivity of the test. Very interesting illustration of a weakness of statistical significance as an indicator of the presence or absence of a correlation between two variables.

Now, I am as sceptical as the next Humanist about the underlying validity of bibliomancy, but it can be a useful seed for meditation. And you have to admit that this selection bears on how we interpret the world around us as empirical beings. So before I tag the next bloggers, I'm going to share a thought or two triggered by the passage quoted above.

Some say that you can prove anything with statistics. In fact, you can only do that if you are dishonest or incompetent, and your audience does not understand statistics. None of us can force every number-quoting ideologue out there to be both honest and competent, but the critical skills needed to discern good statistics from misleading ones are within the reach of anyone who wants them.

Now who's next?

Dale, at Meming of Life
Hemant, the Friendly Atheist
Hugo, who Thinks Too Much
Sean, of the interesting Shore
Kathryn, and her Mindful Life

Belated birth link

Our baby girl, Kaia, is now four months old. When she was born, I promised you, my faithful reader, that I would wax on at length about the strength and beauty of the event.

Time has rushed on, and events have intervened. I am still not able to sit down and write what deserves to be written. But I just checked an old thread at Ben Goldacre's Bad Science forums, and realized that I could give the illusion of writing more by pointing out what I had written some time ago.

So here it is, a link to the childbirth/homebirth discussion.

Have you heard of Jacob Bronowski?

To my shame, I had not until I checked my blog reader this morning.

Yesterday would have been his 100th birthday, and it was marked on both The Panda's Thumb and Pharyngula.

According to Wikipedia, Bronowski's television series The Ascent of Man was part of the inspiration for Carl Sagan's series Cosmos.

What really got me - made me an instant fan of Bronowski - is this clip from The Ascent of Man that was linked on The Panda's Thumb:



It's moving. It's insightful. It's brilliant. It's heart-wrenchingly human.

Scientific thinking at breakfast

I recently read the following on a pack of Kellogg's Fruit 'n Fibre:

FACT: Families who eat cereal for breakfast each day are less likely to be overweight than those who don't.

I get the impression from their website that Kellogg's is more responsible than many companies in making appropriately backed-up claims, so the following is not meant as an attack on them.

However, the scientist in me couldn't help digging into what exactly they're claiming here.

The study that backs up their claim (I assume there was one) was probably designed around a chi-square analysis - they counted the families that fall into four groups - A, B, C, and D - as in the table below:

The "FACT" quoted above amounts to the following:And a bit of simple algebra gives us the following (mathematically-equivalent) espression:This second expression could be read as follows:

FACT: Families that are overweight are less likely to eat cereal for breakfast each day than those who aren't.

So we have the same data, the same facts, but two subtly different statements. One suggests that eating breakfast helps prevent people from being overweight (perhaps even reverses it). The other suggests that being overweight leads people to avoid eating cereal for breakfast.

I could discuss the linguistic distinction between entailment and implicature; I could go on about the role of statistics and the need for precise language.

But remember, this was just a cereal box.

Mainly, this was simply a diverting exercise in taking something you might see every day, and using reason and some straightforward math to dig down into what's really being said.

Humanism, and the empirical skepticism I practiced above, are not just things I bring out when confronted by outrageous pseudoscience or millennial zealots. They are part of how I see the world, day to day. They colour how I react to everything, from ads on cereal boxes to stubbing my toe to deciding what to get Deena for Valentine's Day.

Student group = blog factory?

I don't know what it is - something in the Edinburgh water perhaps - but our student group (The University of Edinburgh Humanist Society) keeps producing new humanist blogs.

I was first - the Friendly Humanist.

Then came This Humanist.

Next, The Not-Quite-So-Friendly Humanist showed up.

And most recently, That Humanist has joined our ranks.

It's tempting to make some cynical comment about our tendency to use transparently derivative blog names, but fans of the Friendly Atheist might have me up for hypocrisy, so I'll just keep quiet.

Anyway, check them out. We're all different. Get a feel for what young humanists in Edinburgh are up to, and how we see the world.

Oldest Milky Way star

Only 500 million years after the Big Bang, a star was born that would later be part of globular cluster NGC 6397, 7200 light years from Earth, which contains of around 400 000 stars. This star, perhaps not the oldest in the galaxy, is the oldest one so far observed in the Milky Way galaxy: about 13.2 billion years old. That's January 14th for those following the Cosmic Calendar. Celebrate the birth of this early-starting star, HE 1523-0901, by getting up just after dawn on Monday.

Note that the galaxy itself didn't start to form from these early stars (and various other bits and bobs) until between 6.5 and 10.1 billion years ago - no earlier than April 6th in our Cosmic Calendar.

I would like to appeal to any experts or students in astrophysics or cosmic evolution to help me populate this sparse end of the Cosmic Calendar with interesting events in the early universe. For now, I'm digging up the occasional tidbit from Wikipedia with my own meagre understanding of the cosmos - a more systematic set of dates would be very welcome.

International Year of Languages

I'm in the middle of analysing data, so I can't talk long. Just wanted to mention that 2008 has been designated the International Year of Languages by the UN General Assembly.

If I were to talk about this, I hope I would say something like what This Humanist says.

I'll also take this opportunity to explicitly list all of the linguistics-related blogs I now know of (let me know if I'm missing any):

• This Humanist
• Ninety-six and ten
• Language log

Part of me is tempted to point out that linguistics is not immune to anti-science creationist foolishness. Another part of me is delighted that language origins are interesting enough that even pre-scientific and anti-scientific thinkers want in on the action.

And another part of me want to use this link-heavy excuse for a lazy post to point you to more reliable sources of information on how languages actually change and diversify. It's a fascinating process, in many ways analogous to species change (and in many ways not analogous). I wonder if demonstrating the observed, documented "speciation" of languages as a result of cumulative "micro-evolutionary" steps would help some of the more honestly-deluded creationists accept the parallel phenomena in biological evolution?

Oh, well.

Here's one last link for today - food for thought for those of us who are tempted to react viscerally instead of rationally when we encounter language change in our own community.