Set the value of alpha. Alpha represents the percentage of error you are willing to make in terms of rejecting a null hypothesis that is actually true. Most statisticians set alpha to 0.1, 0.05, 0.025 or 0.01. The smaller alpha is, the more likely you will safely reject the null hypothesis but the harder it will be to reject it.
Compute the degrees of freedom, or df. The formula for df is always your sample size, n, minus one. For example, if you have a study involving 14 patients, then your df for a t-test would be 14 - 1 = 13.
Calculate your t-critical value according to your alpha level and df. Where the two intersect on the t-distribution table is your desired t-critical value for a one-tailed test.
Interpret your t-critical value as either positive, if your alternative hypothesis is greater than the assumed value, or negative, if your alternative hypothesis is less than the assumed value.