Climate report too quickly embraced by journalists:

So says my latest media analysis column in the Rocky Mountain News, which criticizes the press for its overly credulous reporting of the latest output from the UN's Intergovernmental Panel on Climate Change. The column also looks at media coverage of a bill to mandate HPV vaccines for 6th grade girls; the factoid that only 2% of rape accusations are false; and the lingering influence of Michael Bellesiles on "The Mini Page."

dearieme:
I suspect that there is almost no common accusation known to man with a falsehood rate as low as 2%.
2.10.2007 1:59pm
PersonFromPorlock:

...my latest media analysis column... criticizes the press for its overly credulous reporting of the latest output from the UN's Intergovernmental Panel on Climate Change.

It isn't the journalist's job to critique press releases, only to repackage them as 'news'. You're expecting a pig to whistle.
2.10.2007 2:28pm
Harry Eagar (mail):
Kopel's reference to Paul Reiter and a Google search for 'Paul Reiter malaria IPCC' produces a paper that is enlightening about a little difference of opinion I had with Anderson here last week.
2.10.2007 2:42pm
Elliot Reed:
What is the percentage of false accusations supposed that are false supposed to be a percentage of, and how is "false" operationally defined? That is, are we talking about cases where there was no rape or cases where there was (or might have been) a rape but the wrong person was identified (hardly surprising given the well-known problems with eyewitnesses)? And are we operationally calling an accusation "false" if there is insufficient evidence to show guilt or only if there is sufficient evidence to show innocence? By what standard of proof?

And, for what matter, what's a "rape"? Are we using the legal definition of "rape," a feminist definition of "rape" (which would consider virtually no mistakes as to consent reasonable) or an antifeminist definition of "rape" (which would consider spousal rape impossible or virtually impossible)? Do we count non-rape sexual assault crimes? If someone is accused of legal rape but turns out to be innocent of legal rape and guilty of legal "sexual assault," does that make the accusation false or true?

Without knowing the answers to these questions any statistic presented is meaningless. It would also be interesting to know how the statistics for rape compare for the statistics for other crimes. I bet there are few cases of breaking and entering where no breaking and entering occurred by anyone, but scads and scads of cases where the wrong person is accused.
2.10.2007 2:48pm
JohnAnnArbor (www):

It isn't the journalist's job to critique press releases, only to repackage them as 'news'.

True enough for Greenpeace and PETA news releases. Those by industry, however, that's a different matter. They have an AGENDA, don't you know.
2.10.2007 3:51pm
Viscus (mail) (www):
I wonder why journalists don't take "doubts" raised by right-wing think tanks more seriously.
2.10.2007 4:49pm
Eli Rabett (www):
Care to back up your claim that governments pick those who contribute to the IPCC reports?
2.10.2007 6:13pm
Tim Lambert (mail) (www):
Kopel touts the Fraser Institute analysis. This is not a good idea. Real climate scientists have taken it to pieces here.
2.10.2007 7:29pm
David M. Nieporent (www):
I wonder why journalists don't take "doubts" raised by right-wing think tanks more seriously.
I don't.
2.10.2007 7:48pm
markm (mail):

And, for what matter, what's a "rape"? Are we using the legal definition of "rape," a feminist definition of "rape" (which would consider virtually no mistakes as to consent reasonable) or an antifeminist definition of "rape" (which would consider spousal rape impossible or virtually impossible)? Do we count non-rape sexual assault crimes? If someone is accused of legal rape but turns out to be innocent of legal rape and guilty of legal "sexual assault," does that make the accusation false or true?

The false accusations of concern are ones made to police officers and courts, so the only definition that should matter is the legal one. If you make such a serious accusation, you should be clear about that. If the accusation is based on "I felt psychologically pressured" or morning-after regrets, it's a false accusation. If you had sex (consensual or not) with five men, but none of them were present at the time or place you say you were "raped", it's a false accusation.
2.10.2007 9:56pm
Justin Wilson (mail):
Was this like your "overly credulous" reportage that that lame-ass study you cited last week proving ("conclusively" is the word you used, I think) that the NRA influenced the outcome of elections when the data cited (two elections from a decade ago) actually only showed an influence in 5 of the 18 races.)

I notice you never responded to those who pointed out these deficiencies. Or - if you did - it was several days later.

So is your mode to just drop b.s. statements and never bother to clean up after yourself?
2.10.2007 10:04pm
Elliot Reed:
markm - Counting a case where rape is reported but the offense committed turns out to have been "sexual assault" as a false report might be be seriously misleading. In any case, I'm less interested in how the statistic "should" be defined than in having a clear definition of what the statistic presented is supposed to mean. The critical issue there is probably the standard of proof for determining that a report is false.
2.10.2007 10:29pm
A. Zarkov (mail):
Reporters are supposed to vet their stories. But when it comes to some topics they act like shrinking violets. They seem to like the UN and print its statements with no follow up. African AIDS is a good example. The MSM does not tell its readers that the AIDS prevalence figures UNAIDS puts out for sub-Saharan Africa are likely exaggerations. How many readers know that they don’t enumerate AIDS cases, and they don’t even enumerate positive HIV cases. In general when they identify someone as having AIDS it’s based on symptoms not diagnostic tests. Symptoms easily confounded with malaria. Their AIDS prevalence figures are not even based on area sampling, but a small number of non-representative surveillance samples from a smattering of pre-natal care centers. Then they run that data through epimodel, a really crude algorithm, to estimate the number of AIDS cases. Epimodel is described here. But to fully appreciate the limitations, and assumptions of this technique, the interested person should read this article from the Bulletin of the World Health Organization: Modelling HIV/AIDS epidemics in sub-Saharan Africa using seroprevalence data from antenatal clinics.

You would think this important subject would get a little more critical coverage. Can we really trust the UN on Global Warming?
2.11.2007 1:23am
J. F. Thomas (mail):
Sheesh, talk about the pot calling the kettle black. This is from a man who accepts without question every poorly designed and piece of anecdotal evidence that "proves" ownership of firearms lowers crime and restrictions on the same endangers law abiding citizens and threatens liberty and is the proximate cause of genocide.
2.11.2007 12:07pm
A. Zarkov (mail):
You have me confused with someone else. I might have commented critically on studies proving gun ownership causes excess crime, but not studies showing the opposite—for example Lott. But even Lott’s dubious regressions are a far cry from what UNAIDS has done with epimodel. They can’t even report a confidence interval. Moreover epimodel takes subjective inputs. But don’t take my word for it, simply read the WHO bulletins.
2.11.2007 3:43pm
Harry Eagar (mail):
Eli, from the Paul Reiter statement I mentioned:

33. When I contacted IPCC personnel (at the Meteorological Office in Exeter) to see whether my nomination had been accepted, I initially received the message: "The IPCC received over 2000 government nominations during this process and most, such as yours, were of a very high standard. Unfortunately the IPCC Working Group Two Bureau did not pick you to be an author, although all nominations were scrutinised and assessed".

34. I replied with a question about the two Lead Authors that had been selected: "It is often stated that the IPCC represents the worlds top scientists. I copy to you the bibliographies of (the two lead authors), as downloaded from MEDLINE. You will observe that (the first) has never written a single article, and (the second) has only authored five articles. Can these two really be considered "Lead authors" with experience, representative of the world's top scientists and specialists in human health?"

35. I also pointed out that one Lead Author is a "hygienist", the other is a specialist in fossil faeces, and both have been co-authors on publications by environmental activists. I received the reply: "The selection criteria for IPCC Authors are defined in the "Principles and Procedures Governing IPCC Work" available on the IPCC website at: http://www.ipcc.ch/about/procd.htm (These `Principles and Procedures' have been discussed, amended and agreed by Governments at several IPCC Plenaries)".

36. I pursued the question further, asking: (1) Who selects the Working Group/Task Force Bureau Co-Chairs? (2) Who are the Working Group/Task Force Bureau Co-Chairs for Group II, Health Impacts? Where is the Working Group/Task Force Bureau? (3) What are the criteria they use for identifying appropriate experts?

37. I received two replies, the simplest of which read: "Thank you for your continued interest in the IPCC. The brief answer to your question below is `governments'. It is the governments of the world who make up the IPCC, define its remit, and direction. The way in which this is done is defined in the IPCC Principles and Procedures, which have been agreed by governments. Please refer to my emails of 2 and 3 September for details on how to access that information".

38. In all the rules that were quoted, there was no mention of research experience, bibliography, citation statistics or any other criteria that would define the quality of "the worlds top scientists".
2.11.2007 4:05pm
J. F. Thomas (mail):
You have me confused with someone else.

I was referring to our good friend Dave Kopel, not you.
2.11.2007 4:11pm
karl (mail):
The Heritage Foundation points out that the recently released UNIP on Climate Change document is just a summary of the forthcomming report and was written by UN bureaucrats, not scientists. The full report with scientific assessment will not be released for months and full scientific debate on the report has not begun. So much for journalistic integrity.
2.11.2007 6:40pm
Tim Lambert (mail) (www):
karl, the Heritage Foundation is not telling the truth. The summary was written by the scientists.
2.11.2007 8:32pm
David M. Nieporent (www):
karl, the Heritage Foundation is not telling the truth. The summary was written by the scientists.
Tim is eliding the truth. The government participants got to edit the report.
2.11.2007 9:19pm
David M. Nieporent (www):
Quoting from the UCS website describing the process:
Government representatives are not permitted to edit these book-length reports [that is, the actual IPCC product]; in the end, it is the authors who bear the sole responsibility for the content of their chapters.

However, government representatives do participate in the line-by-line review and revision of the much shorter summary for policymakers, or SPM, for each technical report. The SPM is written by the working group’s lead authors, reviewed in two stages by technical experts, and finally by government representatives before being accepted at the working group’s plenary session. Each SPM is released separately over the course of several months.
Emphasis added.
2.11.2007 9:23pm