fish wrote:Interesting Psyber that you are happy to dish out insults to the worlds climate scientists but you complain when I call your failed attempt at climate science a monumental blunder.
I was talking about how "research" in all fields can be biased by the pursuit of grants and mentioned in a few posts controversy about such activities in medical research.
I was not insulting climate scientists by suggesting that only climate scientists were subject to bending the rules of science under the pressure of "publish or perish" as you seem to be implying here.
The article brief I quote below gives an example of the sort of thing I was referring to:
Bias threatens brain volume studies: Arch Gen Psychiatry 2011; 342:d682
The number of significant positive results found in the literature on brain volume abnormalities is too big to be true, a large meta-analysis has concluded.
After evaluating 41 meta-analyses on the topic, researchers found the number of positive results was almost double what would have been expected based on power calculations. This was suggestive of strong biases in the literature, including selective outcome reporting and analyses reporting, the study authors said. “Such bias threatens the validity of the overall literature on brain volume abnormalities,” they wrote in the Archives of General Psychiatry.
They suggested excess significance might be due to unpublished negative results, or to negative results being turned into positive results through selective exploratory analyses. For example, the typical investigation of brain volumes was likely to measure by default the volume of multiple brain structures. Because of multiple comparisons, most investigations would have at least one positive result to report, even if this was only a chance finding. Although it was evident the literature on brain volume differences was probably subject to considerable bias, this did not mean that none of the observed associations in the literature were true, the study authors said.
“It should be acknowledged that some meta-analyses may be more affected by bias than others and that some may be totally unbiased,” they wrote. “However, the average level of bias is probably large, and steps should be taken to remedy the situation”. Such steps should include, besides the use of newer technologies: the adoption of large multi-centre studies; the standardisation of definitions, outcomes and analyses; and the registration of pre-specified protocols for these studies. For most of the examined brain structures, definitions should be consistent. Significance testing should not be used as a criterion for publication, and journal editors could emphasize the need to make the full data and protocols available, as in other research fields.
“After more than 25 years of research in this field, further progress requires stronger guarantees of reliability for ensuing results,” they concluded.
By the way, I raised the question you have objected to my even asking here - the one about the comparability of flask air and Ice Core bubbles - with Mike Young, of the Environment Institute at the University of Adelaide, while talking to him at a function there. He said he did not actually know the answer himself and referred me to John Tibby.
John was quite helpful and has convinced me in his polite, non-abusive, manner, and his preparedness to send me a paper I could read myself, that there is a significant human contribution to current CO2 levels on top of the current cyclical peak. Politeness and providing references worked where bullying didn't.
John Tibby wrote:Thanks for your email and the very interesting question. Mike is perhaps too kind when he says I am the right person to ask, but nevertheless I know the literature through both teaching and research.
Essentially, the ice core records and those measured from flasks (including at Cape Grimm and Mauna Loa) can be compared and then “stitched together” through measurements in firn and ice at sites with high rates of accumulation (such as that the attached paper). The high resolution measurements of recent firn and ice can then be tied to the longer lower resolution records such as that from Vostok (through comparison of the Law Dome and Vostok ice core CO2 data). This study demonstrates that there are no effects associated with fractionation that can account for recent high CO2.
