Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
significance of data
Message posted by David (via 137.99.33.89) on December 28, 2001 at 12:43 PM (ET)
I have a set of 6 data points with associated standard deviations. On an x-y plot, one appears to be much lower than expected. I need to show that this difference is significant. I have done a 2nd order poly fit to the other SD error bars, and the SD of the point is outside this fit. Questions: (a) is this kosher, and (b) what's the right way to do this?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: significance of data
Message posted by JG (via 128.8.22.30) on December 28, 2001 at 6:51 PM (ET)
Re: significance of data
Message posted by JG (via 128.8.22.30) on December 28, 2001 at 6:51 PM (ET)
How can a point have a SD ?
Re: significance of data
Message posted by David (via 137.99.33.89) on January 8, 2002 at 12:07 PM (ET)
Pardon the slip... To be more precise, I should have said, "points with associated standard deviations." Each represents three measurements. Does that help?
Re: significance of data
Message posted by Jack Tomsky (via 12.144.103.66) on January 8, 2002 at 5:29 PM (ET)
You can use Bartlett's test to test for the equality of the six standard deviations.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.