Negative t-test scores

[vj...@hotmail.com]

What does it mean if my t scores come out negative? Is that ok? or Did i do
something wrong? I used spss 7.5 to generate it. Thanks for your help.

Joy

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

[Kent Campbell]

vj...@hotmail.com wrote:
: What does it mean if my t scores come out negative? Is that ok? or Did i do

: something wrong? I used spss 7.5 to generate it. Thanks for your help.

: Joy

Hi Joy -
negative t statistics are OK, don't worry about it. (Don't
forget that the t test generates a t statistic and not a T score - my
students sometimes mix up T and t.)

Best wishes,
Kent.

[vj...@hotmail.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.

[KMitch1988]

It's ok to get a negative t score. Although you did not mention the test you
were doing, it's usually applicable in the case of a two tailed test of
hypothesis. Here is an example. Let's say you were testing the hypothesis:
Ho: m= 300
Ha: m not =300

Let 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.

[suresh...@gmail.com]

[Rich Ulrich]

On Thu, 19 Apr 2018 14:00:19 -0700 (PDT), zhangy...@gmail.com
wrote:

>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]

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

[Rich Ulrich]

On Tue, 27 Oct 2020 07:36:41 -0700 (PDT), Mohammad Rasel
<rasel.ka...@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