Ordinal Regression: Data to Order

On Thursday, Eric Michael Johnson tweeted a request for help:

Pop Quiz: You have 3 nominal categories of ordinal data collected at 3 separate time intervals. What statistical test is most appropriate?

Once I had past my usual bee in the bonnet about p-values, I responded with The Correct Answer, but fully explaining what it meant takes more than 140 characters. Josh Rosenau had also responded suggesting an repeated measures ANOVA, which is either wrong or sort-of right, depending on how you look at it.

Continue reading

Posted in Statistics | Comments Off on Ordinal Regression: Data to Order

Some Important and Weighty Stuff in Statistics

(note: to create maximal puzzlement I have cross-posted this at my other blog)
Right, now let’s talk about methods for stochastic numerical integration, in particular calculating importance weights. This is a technique that is probably underused, largely because…
Why are you staring at the screen like that?
OK, OK. I’m sorry for putting up such a technical post,but I was going to write this up for a colleague, and then thought that there might be 3 other people in this world that would also find it interesting.

Continue reading

Posted in R, Statistics | 2 Comments

Pleasing News

Last year, my student Crispin defended his PhD thesis. The last papers from his thesis have been appearing. This is good news, obviously, but it gets better.

Continue reading

Posted in Science Blogging | 1 Comment

More Ideas About Ideas in Ecology and Evolution

A couple of years ago I blogged about a new journal, Ideas in Ecology and Evolution, and it’s experiments in the reviewing process. I was sceptical then, but happy to be shown wrong: I think we need these experiments to find out how . For this year, Ideas in Ecology and Evolution (IEE) is starting a new experiment, and this time they seem to have jumped the shark.

Continue reading

Posted in Science Publishing | 11 Comments

As sick as a Beast

I can’t remember if I’ve had cause to mention this before, but The Beast is an Australian. This may explain his reaction to the news last week of England winning The Ashes series 3-1: he was throwing up the following morning. Of course an English cat would not have reacted in such an extreme way. Partly because of a stiff upper lip, but largely because it would have got used to results like that as a kitten.

Continue reading

Posted in Silliness, Statistics | 2 Comments

Banned! (no, not me)

Oh dear. The Chinese translation of Andrew Gelman’s book on hierarchical models and Bayesian analysis has been banned, due to “various politically sensitive materials in the text”. He must be so proud.
I’ll check the text tomorrow to see what might be so sensitive – perhaps they don’t like instrumental variables. But until then feel free to speculate, in an appropriate manner.

Continue reading

Posted in Silliness | 4 Comments

Yes.

Linda Lin just blogged about the new Google Analytics we have on NN. The most popular keyword (which I assume mean search term) is “Am I wasting my time?”. Irrespective of the message of the post it leads to, the answer is clearly “yes”.
Good. We’ve got that sorted out now.

Posted in Silliness | 3 Comments

Yay, a banner!

I finally worked out The Secret to putting a banner up (it would have been quicker to read the instructions, I suppose). So The Beast is now proudly at the top of the blog, where he feels he belongs.

Continue reading

Posted in Meta | 9 Comments

The Hilarity of Homeopathic Housing

Thanks (I think) go to Rob Hinkley for pointing out John Benneth’s blogpost “IN ONE YEAR: Homeopathy could have helped. It’s the bloggy equivalent of walking around with a “Kick Me” label sellotaped to your back.

Continue reading

Posted in Silliness | 21 Comments

Blame the system!

Sometimes the system you’re working on just doesn’t work the way you hope. So what do you do? Write a paper in PNAS blaming the system? Well, a paper has just appeared in PNAS with the title “Bimodal gene expression in noncooperative regulatory systems”. Damned regulatory systems – they should be water-boarded.

Continue reading

Posted in Research Blogging, Silliness | 3 Comments