On Thu, May 17, 2012 at 6:36 AM, David Roberson <[email protected]> wrote:
A characteristic of HID lights is that they begin to cycle on and off as > they approach the end of life. Time to replace those Slider lamps. > > Dave > The human mind loves to find patterns, even when they don't exist. Certain industries appear to be predicated on identifying patterns that don't exist. Statistics is a tool we've developed to protect ourselves from our flourishing imaginations. It helps us to get beyond mere hunches about whether there is a pattern. Without it, we are left to our own devices when it comes to understanding a patchwork of sporadic observations, and our creative interpretation of events can easily gets the better of us; without our taking careful measurements and then determining whether a hypothesis is statistically significant, I'm sure we are altogether helpless to know whether something is real or not when the data are sparse. I will go out on a limb and suggest that it is largely a well-trained distrust of personal experience that characterizes the scientific approach to things (I'm no scientist, so I can only offer conjecture here!). The question: do some people influence street lamps? The answer: let's take a bunch of measurements and run a regression against them. That said, I wouldn't be surprised if scientists periodically throw out statistically significant patterns that don't accord with their own understanding of things -- perhaps ESP is one. Eric
