Two Sample Tests - Agricultural Statistical Data Analysis Using Stata - page 68

This results in the following output:
One-sample t test
-----------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf.Interval]
---------+-------------------------------------------------------------------
fieldy~d | 120 93.82417 2.468546 27.04157 88.9362 98.71213
-----------------------------------------------------------------------------
mean = mean(fieldyield) t = 15.7275
Ho: mean = 55 degrees of freedom = 119
Ha: mean < 55 Ha: mean != 55 Ha: mean > 55
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000
From this output, we can see that the mean from this experiment
(93.8 lbs/plot) is considerably higher than the average yield of 55 lbs/plot.
The t is the calculated t value, which is used to determine a probability
of statistical significance. At the bottom of this output table are three
listed probabilities (Pr). The first indicates there is no chance the mean is
lower than 55. The second indicates that there is a significant difference
between the calculated mean and 55, while the last indicates the calcu-
lated value is significantly greater than 55. This may seem unimportant
in this case, but there are cases where the researcher is specifically inter-
ested in whether a value is above or below a specified value.
The two-group t-test determines if there are differences between
two groups. Open the file simplepumpkin.dta and enter the command
ttest yield , by ( variety )
This command compares two means using variety as a grouping vari-
able; in this case, two different pumpkin varieties. The yield variable
represents the yield from four plots from each variety.
The output is slightly different than displayed with the previous use
of this command, but will give information similar to the above output.
Paired t test
-----------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+-------------------------------------------------------------------
normal | 4 286.225 41.15072 82.30145 155.265 417.185
orange~l | 4 356.525 132.1471 264.2942 -64.0261 777.0761
---------+-------------------------------------------------------------------
diff | 4 -70.3 159.65 319.3 -578.3775 437.7775
-----------------------------------------------------------------------------
mean (diff) = mean(normal - orangebull) t = -0.4403
Ho: mean (diff) = 0 degrees of freedom = 3
Ha: mean (diff) < 0 Ha: mean(diff) != 0 Ha: mean(diff) > 0
Pr(T < t) = 0.3447 Pr(|T| > |t|) = 0.6895 Pr(T > t) = 0.6553