Paired t-test is used to compare the mean of a certain characteristics before and after intervention. The data before intervention is called pre-test data while the data after intervention is called post-test data. That characteristic should be numeric in nature. For example, we can compare the mean systolic blood pressure before and after the intervention. The following steps clearly illustrate how to calculate paired t-test using SPSS.
1. First of all, we should have the dataset with our desired variables of interest. For example, we can have a variable named "systolicBP_Pre.sav" which has the pretest data for systolic blood pressure. Similarly, we can have a separate variable named "systolicBP_Post.sav" which has the post-test data for systolic blood pressure. Now, we want to calculate whether there is effect of an intervention (e.g. anti-hypertensive drug) in reduction of blood pressure and test whether that reduction is statistically significant.
analyze, compare means and paired samples t test
4. Drag the pre-test variable in variable 1 box and post-test variable in variable 2 box.
5. If you want to change confidence intervals, click on options, otherwise leave it as it is (The default CI is set to 95%).
6. Click on continue and and click on ok, then you will see the output showing the comparison of systolic blood pressure before and after intervention.
7. Interpretation of the output:
From the given output, you can see that the mean systolic blood pressure before intervention was 170.270 and after intervention, it was 117.297. It also calculates the correlation coefficient which shows the value of 0.135 with the P-value of 0.425 which means that the two sets of observations are not related.
Similarly, the table for Paired samples t-test shows that the mean difference in systolic blood pressure is 52.97 and the calculated t-value is 21.128 with 36 degrees of freedom. P-value of 0.000 means that the given P-value is less than 0.05 that indicates that the difference in the systolic blood pressure because of intervention (i.e. anti-hypertensive drug) is statistically significant.
The following video explains the simple steps on how to perform paired t test using SPSS.
Good luck and see you all in the next tutorial!!