I must say that it affects one’s attitude to the data (and I have to watch out for this), as one inevitably is more critical of data that you have to fight to get.

I find that people who are willing to share their data with me at least believe their own analyses.

Those who won’t share their data, or are very reluctant to share, tend not to believe their own work will stand up to careful an analyses.

Speaking loosely, those of us who work with data do statistical tests to be (for example) 95% confident that a pattern this strong is not just the result of random variation.

When someone won’t share data, IMO their statistical analyses no longer meet the 95% confidence standard. In my experience, those who won’t share data are likely to have made major errors (or worse), so their work no longer meets even a 50% confidence standard for me, let alone a 95% confidence standard.

James Lindgren
Northwestern University

But all that aside, I notice you don’t baulk at using the word ‘information’ in the same, singular, group sense you’re objecting to with ‘data’ when trying to explain why data should be plural, Chris. Surely, if ‘information’ in the scientific arena always consists of multiple elements, the word ‘information’ should always be treated as a plural. These information…?

It’s good to wield language precisely; it is one of our most valuable tools after all. But we’re supposed to use language rather than let language use us. To my mind, Steve is clearly referring to a group when he writes ‘this data’; but it’s one single group (even though there are many things in it). Which is why it has a singular article.

Are you french? I’m from Luxemburg and we speak both, french and german

Yes, you are right. Maybe Mr Rahmstorf caught a cold, or he was angry that more than 70% of the posters were more or less sceptic about climate science. Some weeks ago he explained to a true AGW-believer, that realclimate started the same way, and had still become what it is today.

But, it just didn’t work in Germany. I don’t know why. Maybe because the blog was from Mr Rahmstorf?

Best regards

Eddy

Yet we do say museum/musea etc. So the Englishdo honour the Latin forms regularly. The answer of course is language is not logical. By the way, even your linguist friends would not approve of “datas”.

Steve,
Could you do me a favour and treat the word “data” as a word that stands for multiple bits of information especially if you refer to bristlecone data (as you know the Latin word data is the plural form of datum). So when you say ” …..any reasonable selection criteria governing Tornetrask or Yamal or bristlecones, this data wasn’t used in any of the AR4 multiproxy studies, nor was it used in Mann et al 2008″, it hurts my eyes to read “this data” instead of the correct “these data”. Especially in the scientific arena information always consists of multiple elements and the word data should always be treated as being plural.

It is already bad enough that the English language mistreats the Latin word “visum” by saying “visa” when they refer to a single entry document. Even worse is the usage of “multiple visas”. That’s like saying “datas”. The English abuse the origin of Latin words also when they say the “the media is” instead of the the “media are”. A single media doesn’t exist; it is by definition plural.

Sorry for sounding pedantic but it has bothered me for years and I feel relieved to get it off my chest. A blog is indeed the ideal medium (not in a clairvoyant sense!!) for people to vent their frustrations.

Watt’s Up With That is way ahead, with Pharyngula second, CA third, Bad Astronomy a close fourth, and RC in distant but still-viable 6th place.

RC encourages its readers to vote, but cautiously dismisses the outcome should it favor a wrong blog: http://2008.weblogawards.org/polls/best-science-blog/.