Discussion:
Negative t-test scores
(too old to reply)
s***@gmail.com
2017-04-16 13:01:02 UTC
Permalink
your level of significance (alpha) = .05. Let's also asume that there are
8 observations.
The degrees of freedom would be (n-1)=7. Your tabled t value for alpha =.05
with 7 degrees of freedom is 2.365. You will therefore reject Ho if your
calculated
t > 2.365 or t < -2.365.

If your t value came out to be -2.97, that means it is in the rejection region
and you would therefore reject Ho.

Rationale: It depends on how you define your rejection and acception region.
z***@gmail.com
2018-04-19 21:00:19 UTC
Permalink
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.
Rich Ulrich
2018-04-20 17:45:14 UTC
Permalink
Post by z***@gmail.com
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
Mohammad Rasel
2020-10-27 14:36:41 UTC
Permalink
Post by Rich Ulrich
Post by z***@gmail.com
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?
Rich Ulrich
2020-10-28 05:05:19 UTC
Permalink
On Tue, 27 Oct 2020 07:36:41 -0700 (PDT), Mohammad Rasel
Post by Mohammad Rasel
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
Loading...