Posts by linger
Last ←Newer Page 1 2 3 4 5 Older→ First
-
There is a difference between journalism and science. The National Journal article presents a journalist's argument, not a scientific argument.
In my summary of the article above I fairly deliberately gave it the benefit of the doubt, in focussing on the substantive points that directly attacked the Lancet paper. It would be too easy to identify points that are less relevant to a scientific argument, but (applying the article's own logic) undermine the article's own credibility.* It shows notable political bias (sure, an ad hominem -- but in this case it directly leads the authors to glaring logical flaws, as below).
* Risible argument from authority (it is implied, more than once, that the Lancet study has no credibility simply because George W said so).
* Astounding scientific howlers (e.g. someone is quoted as saying that the environment of Iraq had been spoiled with effects lasting for "35 billion years". Bearing in mind that the Earth is only 4.6 billion years old, and will in all likelihood be burnt to a cinder within a similar period, this shows a lack of care to accuracy that counsels caution in assessing the "factual" claims made against the Lancet paper).
Does any of this prove that the article's attack is unfounded? Actually, no.
But by the same token, the article's attack does not prove that the Lancet study is unfounded. As I said above, it raises questions, but does not definitively answer them.I would add that journalists are generally not all that trustworthy in reporting on science. (I am not claiming any intent to deceive. Simply that they're not experts, and it shows). For example:
Bell (1989) conducted a study of newspaper reports concerning global warming, by the simple method of sending copies of articles back to the sources quoted and asking them to comment on any inaccuracies. He found that
"Sources rated 29% of stories absolutely accurate, and 55% slightly inaccurate, with 16% in the higher inaccuracy levels [...] Scientific/ technical inaccuracies were present in about a third of stories ﾑ technical terms misused, wrong figures given, scientific facts confused. Non-scientific inaccuracies (such as mis-spelled names, misnamed organisations and wrong dates) also occurred in about a third of the stories, as did misquotations. About a quarter of stories had significant omissions or exaggerations."
(Bell, Allan (1991) The language of news media. Oxford: Blackwell. p217)
-
the article on the Lancet study [...] is a devastating destruction
Not exactly.
The article raises a great many questions (some admittedly well worth asking) -- but it does not provide definitive answers. That must await detailed peer review, and attempts by other researchers to replicate the Lancet study's findings.The main claims made in the article are:
(1) The authors and researchers show political bias.
[By itself, this is simply an ad hominem.](2) The sampling methodology used is possibly unrepresentative
(e.g. favouring main streets where car-bombs were more likely).
[This is not clearly demonstrated in the article, but may be worth following up.](3) The data collectors were not effectively supervised,
and working under extreme time pressure and in extreme danger
[Seems a fair comment.](4) As a result, the data collectors may have fabricated some data (the article cites some suspicious details -- unlikely distributions of deaths, missing death certificates, unexpectedly high survey response rates -- in support of this claim).
[But accusation is not proof: this is where replication is needed.](5) Nobody has been given access to the raw data, and therefore (under the circumstances) we should question its existence.
To return to our original topic: only (1) -- the ad hominem -- is potentially relevant to GW. So I have to concur with Kracklite that in the current thread, this is a non sequitur.
-
Of course, it is highly probable that (if the system were to permit it) "no confidence" would win over any of the listed candidates, especially for the Republicans. Which is the likely reason for the much lower Republican turnout -- their choices are pretty much as characterised by Craig above. (Not unlike certain NZ constituencies. I mean, seriously, Banks!?) So the numbers shouldn't be seen as a direct "Democrat vs Republican" vote at this stage.
-
(Erratum: For "exploratory" above read "confirmatory".)
-
Just in case some missed that 50% is one half
Your point being...?
(i) 50% is actually a pretty high proportion of variance to be captured by a principal component.
Depending on how many variables went into the analysis, it might be possible to interpret a PC with only 10% or even 5% of the variance.(ii) climate data is complex; there isn't just one cause, and so we should expect several potentially interpretable factors -- and certainly not all the variance to be captured by one factor (even if all the data going in were hypothesised to be explained by one factor).
PCA is a useful way of extracting the most important contributions from complex data. It's best suited to summarising data that is already hypothesised to have one main cause; that summary dimension should be the first principal component. It is not an exploratory method (it doesn't directly allow you to test hypotheses about the "best possible" factor structure), but it also has certain advantages in that other researchers can replicate the results, and do follow-up analyses, more easily than with other types of factor analysis.
-
Governments should hand out gongs to their political donors (just ignore what we said yesterday). We need a group of public spirited wealthy men to support political parties, to ensure they pursue sound, sensible and responsible policies, and so they don't have to seek the support of ordinary voters instead.
Uh, guys, are you quite sure this isn't intended as sarcasm? Given the miniscule likelihood of the Herald actually retracting their opinion, that would seem to be the most likely interpretation.
-
I have some bad news about the snot fairy (and even the earwax fairy):
it's been done -- albeit published only 2 months ago.Flim-flam Fairies
Alan Katz, Illustrated by Michael Slack
[...]
A tale about how the Belly Button Lint Fairy, Earwax Fairy, Dirty Underwear Fairy, Clipped Toenail Fairy, Snot Fairy, Fart Fairy and others attempt to take over the Tooth Fairy's under-the-pillow enterprise. They try to profit from pliable bogeys, get funds from farts and make dirty underwear pay.... er. actually, I gather that the last-mentioned is already a lucrative sideline in this city. But I hope that's not quite what Katz has in mind.
anyway, the more positive conclusion is that there are some publishers out there willing to venture into areas more suited to children's than parents' sensibilities.
-
oh, booger...
let me guess, the snot fairy's rejection by mainstream society results from it being seen as some kind of bogeyman.
Actually, the user comments on Briggs' book seem equally appropriate here, David:
This book would be of interest to any kid over 7, boys might go for it at an earlier age than girls - although be warned it's not suitable for sensitive parents. It's also ideal for well read teenagers and young adults, who will appreciate the sophisticated humour more.
-
it must be a taste divide, rather than a gender divide, when it comes to snot fairies.
obviously, as signalled by Deborah (quoting daughter) above:
Q. What's the difference between snot and broccoli?
A. Kids don't eat broccoli.Presumably if one were to puree the broccoli and add salt one might get a better reaction? or would it also have to be inserted nasally?
-
After all, our bodies are mass and energy are they not? And, correct me if I'm wrong, but energy never dissipates, does it?
I don't want to be seen as attacking your source of spiritualiuty here, but it may be amistake to try to base it on sciente in this way.
After all, our bodies are mass and energy are they not? And, correct me if I'm wrong, but energy never dissipates, does it?
I don't want to be seen as attacking your sense of spirituality here, but it may be a mistake to try to base it on science in this way.
Mass-energy is conserved. To a reasonable approximation, that means energy is conserved (if not converted to matter).
But this doesn't mean that energy doesn't "dissipate" -- in fact, by default, that is exactly what happens; energy is transformed into less organised forms.Life is a localised reversal of entropy [possible only because it can draw directly or indirectly on localised organised energy sources]. Death ... isn't, as far as can be observed.