Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
For the out-of-band Slashdot experience (mostly headlines), follow us on Twitter, or Facebook. ×
Science

No Higgs Just Yet 190 190

gbrumfiel writes "Last month, scientists reported a number of 'excess events' that could be caused by the appearance of the long-sought Higgs boson inside the LHC. But it looks like they'll have to put the champagne back on ice. New data presented at a conference in India shows no new signs of the Higgs. The signal was probably just a statistical fluctuation."
This discussion has been archived. No new comments can be posted.

No Higgs Just Yet

Comments Filter:
  • Re:Huzzah! (Score:4, Informative)

    by Rising Ape (1620461) on Monday August 22, 2011 @04:58PM (#37171554)

    That's not an accurate way of looking at the statistics, but it's a common mistake (and most likely one I've made myself).

    The confidence levels (such as 99%, 95%) tell you very little directly about the probability that a result is correct. What they do tell you is the probability of getting a false positive - if you were to do 100 separate searches at a 95% confidence interval, you'd get a positive result in about 5 even if there were no Higgs there.

    Given the number of potential places to search and the number of experiments done, you'd expect some false positives at these confidence levels. The standard for claiming a discovery is 5 sigma, which is something like a 99.9999% confidence level.

  • by Lore17 (1318959) on Monday August 22, 2011 @05:28PM (#37171960)

    The Kruger-Dunning effect: http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect [wikipedia.org]. The less you know about something, the more certain you are that your intuitions about it are correct.

When someone says "I want a programming language in which I need only say what I wish done," give him a lollipop.

Working...