http://www.climate-skeptic.
Part of any scientist's work is doing this sort of analysis, and it's done pretty well at the above link. It's a good illustration, not of "why even reasonable people can draw different conclusions from the same data set" (I think the author was being over-generous), but rather of how pseudo-scientific analysis on carefully limited data sets can be used to gin up whatever headline you'd like.
Me? I would've first FFT'd the whole mess and filtered out the seasonal variation frequencies, inverse FFT'd, then started looking for trends. Then tacked on a giant disclaimer about how the results are only preliminary and that many more years' data are needed before any conclusions can be made. But that's just me.
OK (and I spell it "OK" not "Okay," just deal) that may have been about climate change but it had exactly zero political content. Was all about doing math, displaying data, and not letting your much-desired conclusions get in the way of what the data might be able to tell you.
No comments:
Post a Comment