|
|
The t-Test analysis is probably the most well known statistical analysis.
It is simple and easy to understand. The test determines whether measurements in one group are significantly different from the
measurements in the other group. That's pretty much it.
There is one distinction to make: whether you have two independent or dependent samples/groups. If you take
measurements from two distinct groups, for example, 50 people chosen at random receive a treatment and 50 others don't, then you have
two independent samples. On the other hand, if you took measurements on a person before and after treatment, then those measurements are
clearly not independent. Therefore, you have a "paired sample".
We will not get into details about the t-test, there is already vast documentation.
In Mineful, you simply select a numeric and a group variable. This group variable may have more than 2 groups, but since the t-test only examines two samples, select
only two groups (or Mineful will scold at you). The output includes: t-statistic, p-values, confidence interval for the difference in means, method (equal or unequal
variances, yes Mineful checks prior if the variances are equal to use the proper method), and a brief conclusion on whether the averages
in the two samples are significantly different or not.
For technical details about the t-test and Mineful, visit our t-test support page.
|
