• Negative t-test scores

    From Mohammad Rasel@21:1/5 to Rich Ulrich on Tue Oct 27 07:36:41 2020
    On Friday, April 20, 2018 at 11:45:22 PM UTC+6, Rich Ulrich wrote:
    On Thu, 19 Apr 2018 14:00:19 -0700 (PDT), zhangy...@gmail.com
    wrote:
    On Saturday, January 23, 1999 at 3:00:00 AM UTC-5, vj...@hotmail.com wrote: >> Dear Kent, Thank you so much for the reply. And yes, i forgot they are called
    t statistic. (sorry about that..:) One more question, should I worry if the
    standard error of the mean is bigger than the mean itself? My thesis adviser
    says i should find out. Thanks.

    Joy

    -----------== Posted via Deja News, The Discussion Network ==----------
    http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own

    It is okay your standard deviation is larger than your mean. This just means that your sample is quite dispersed.

    Did you notice - you are Replying to a message from 1999?

    Your answer ("quite dispersed") is based on the assumption
    that all the scores are positive. When the mean is near zero,
    owing to negative scores, it is very possible to have low
    dispersion when the the SD or SE is larger than the mean.

    --
    Rich Ulrich
    please help me to interpret my result. my t value of regression has negative value.what does that actuallt mean?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Rich Ulrich@21:1/5 to rasel.kabir06968@gmail.com on Wed Oct 28 01:05:19 2020
    On Tue, 27 Oct 2020 07:36:41 -0700 (PDT), Mohammad Rasel <rasel.kabir06968@gmail.com> wrote:

    please help me to interpret my result. my t value of regression has negative value.what does that actuallt mean?

    A negative t-value for a regression coefficient b in

    Y = b*X + C

    says that the predicted value for Y goes down as
    the value of X goes up. If you fit a plot of X and Y,
    the slope is negative.

    --
    Rich Ulrich

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)