Gather data for members of each of the two groups. These data should consist of the value of the quantitative variable (height, temperature, income) for each subject in each group.
Find the mean in each group. Many calculators have buttons for the mean, but you can simply add up the numbers and divide by the number of subjects. Call the mean for one group "m1" and the mean for the other group "m2."
Find the variance for each group. This is much easier if your calculator has a button for the variance, but it can be done by hand by subtracting each value from the mean for that group, squaring that difference and adding the squared values. Call the variance for the first group "v1," and for the second group "v2."
Divide each variance by the number of subjects in that group.
Add the two numbers from Step 4, and take the square root of that sum.
Subtract m2 from m1.
Divide the result from Step 6 by the result from Step 5. This is "t."
Compare "t" to critical values found in a t-table to test its statistical significance. These are in the backs of many statistics books, or online (see Resources). Degrees of freedom is one less than the total number of subjects in the two groups, combined.