- 1). Determine the alpha value you wish to apply to your evaluation of the t-test. Alpha is the percent chance (in decimal form) that you will make a mistake in your conclusion. Specifically, it is the percent chance that you will mistakenly conclude that there is a difference in the two samples, when in fact there is not. A common level for alpha is 5 percent (0.05).
- 2). Determine the "degrees of freedom" of your sample data. Calculate this by adding the number of numerical values in the first sample set to the number of values in the second set, then subtracting 2.
- 3). Look up the critical t-test value on a t-table. Find your degrees of freedom on the left of the table, and your alpha value along the top row. Where the row corresponding to your degrees of freedom and the column corresponding to your alpha meet, you will find the critical t value you will use.
- 4). Compare your t-test statistic (previously calculated from your sample data) with the critical t value found from the table. If your calculated t value is greater than the critical t value from the table, you will conclude that there is a statistically significant difference between the two samples. If not, you will conclude there is no difference.
SHARE
