Paired Sample t-test Using SPSS or R

Get a statistician’s help calculating a paired sample t-test using SPSS or R. On this webpage, we use data located here https://hcup-us.ahrq.gov/db/state/siddist/siddist_discharge.jsp [external site] to compare the number of discharges in all states between 2012 and 2019 using a paired sample t-test and describe the assumption of a paired samples t-test. The provided paired t-test example was performed using SPSS.

A paired t-test assumption is that the dependent t-test compares the means on the same continuous, dependent variable between two related groups. The dependent variable ought to be measured on a scale that is continuous, which means it is measured at the ratio or the interval level, according to the t-test’s assumptions. The other paired t-test assumption is that the independent variable should have two classifications; matched pairs or related groups.

The phrase ‘related groups’ denotes the presence of the same subjects in both groups. In SPSS Statistics, this test is known as the paired-samples t-test. A paired sample t-test was conducted at a .05 significant level to determine if there was a statistically significant difference in the number of discharges between 2012 and 2019 among the states in the region. The findings obtained revealed that the difference in discharges between 2012 and 2019 was statistically significant, t (20) = 8.160, p < .001 (see Table 1). The normality assumption was tested using a histogram displayed with a normal curve. The normality assumption was met (see Figure 1). Additionally, a box plot was included to check for the presence of outliers in the data. There were no outliers found in the data (see Figure 2).