Analyzing two-sample t test

[Gabriel Meynet]

Hello everyone,

I've made a quick research on the internet and the forum and I've not found my answer so here am I:

I've some troubles getting the p-value out of this simple t-test made on stata, comparing the totalscore (gaussian distribution) of two population:

ttest scoretot, by(inf)

Two-sample t test with equal variances
------------------------------------------------------------------------------
Group | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+--------------------------------------------------------------------
med | 218 46.80734 .571411 8.436783 45.68111 47.93357
inf | 290 43.12759 .4978929 8.478811 42.14763 44.10754
---------+--------------------------------------------------------------------
combined | 508 44.70669 .3836415 8.64684 43.95297 45.46042
---------+--------------------------------------------------------------------
diff | 3.679753 .7584321 2.18969 5.169817
------------------------------------------------------------------------------
diff = mean(med) - mean(inf) t = 4.8518
Ho: diff = 0 degrees of freedom = 506

Ha: diff < 0 Ha: diff != 0 Ha: diff > 0
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000


I'm a bit ashamed to ask that question because the answer might be quite simple: would it be the "t= 4.8518", or is it one the three probability below (which I don't really understand by the way) ?
I just need to be sure of that.

I'm already grateful for your attention and time and wish you all a good day !

Gabriel

[Carlo Lazzaro]

Gabriel:
-t- and p-value are linked indeed.
If you're looking for a two-way p-value,

Pr(|T| > |t|) = 0.0000

is the information you need.
As an aside, for the future please post what you typed and what Stata gave you back via CODE delimiters (see the FAQ on that and other interesting posting requirements. Thanks).

[Gabriel Meynet]

Thanks for your reply and sorry for the wrong layout of my post.

Best regards,

Gabriel

[Marcos Almeida]

Carlo gave a very helpful reply, as always.

That said, since you remarked

I've made a quick research on the internet and the forum and I've not found my answer

Your doubt is related to the core-information related to stats notation, not necessarily Stata.

That said, if you search on the Web, having as key words "Stata t test interpretation", you will be shown, among other useful resources, this text.