T-Tests are used to compare the means of two groups of scores. The scores can be from the same group of people obtained at different points in time or from two groups of independently sampled responses. SPSS (Statistical Package for the Social Sciences) is used to calculate statistical tests, including T-Tests. The output from the program will provide the T value (the test statistic or value obtained) and the significance level (the probability the results are due to the independent variable rather than chance) for the comparison.
Click on the data sheet at the bottom of the SPSS screen. Double click on "var0001," which will produce a dialogue box. Type the name of your first variable in the box--for example, test 1--and hit "OK."
Double click on "var0002" and type in the name of your second variable--for example, test 2--then click "OK."
Enter the first score for your first group in the first column, below the first variable, and click "Enter." Continue entering data down the column.
Enter the first score for your second group in the second column, below the second variable, and click "Enter." Continue entering data down the column.
Click on "File," then click on "Save As," and enter the name for your file. Click on SPSS(*.sav).
Click on "Statistics" at the top of the SPSS screen. Select "Compare Means" in the drop-down box. Select "Paired-Samples T-Test."
Select the two variables (test 1 and test 2) while holding down the "Shift" button, and move them to Paired Variables. Click "OK."
See the results in the output window. The output will indicate the T-value (for example, 1.32) and the significance level (for example, .057). In this case, .057 indicates that there are 5.7 possibilities out of 100 that the results are due to chance.
Frequently save your data.
Tips and warnings
- Frequently save your data.