st: about Two independent samples t-test

[rghuang <0034058@fudan.edu.cn>]

if  i wish to test whether the mean for write is the same for males and females.  i use

. ttest write, by(female)

then the result is:

Two-sample t test with equal variances

 

------------------------------------------------------------------------------
   Group |     Obs        Mean    Std. Err.   Std. Dev.   [95% Conf. Interval]
---------+--------------------------------------------------------------------
    male |      91    50.12088    1.080274    10.30516    47.97473    52.26703
  female |     109    54.99083    .7790686    8.133715    53.44658    56.53507
---------+--------------------------------------------------------------------
combined |     200      52.775    .6702372    9.478586    51.45332    54.09668
---------+--------------------------------------------------------------------
    diff |           -4.869947    1.304191               -7.441835   -2.298059
------------------------------------------------------------------------------
Degrees of freedom: 198

 

                  Ho: mean(male) - mean(female) = diff = 0

 

     Ha: diff < 0               Ha: diff != 0              Ha: diff > 0
       t =  -3.7341                t =  -3.7341              t =  -3.7341
   P < t =   0.0001          P > |t| =   0.0002          P > t =   0.9999

from the result,we can say females have a statistically significantly higher mean score on writing (54.99) than males (50.12).

 

as the result say this test should be with equal variances ,so should we use ttest when we don't know if the variances is equal?

 

so whether should we use sdtest to find if the variances is equal?if not equal,should should i do if i want to see whether the mean for write is the same for males and females?

 

 

thank you!

 

 

Shanghai

Fudan University