A small study investigated whether two forms of aspirin differed in their average bioavailability, with bioavailability being measured in terms of the aspirin concentration in urine 1 hour after taking the aspirin. In this study, ten people were randomly allocated to receive either AspirinA or AspirinB. One hour after administering the drug, a urine sample was taken and the aspirin concentration measured. One week later (i.e., after the first aspirin dosage had been fully excreted), each participant received an equal dose of the other aspirin, and once again a urine sample was taken an hour later and the aspirin concentration was measured.

Should the data (i.e., the aspirin concentrations) from this study be analyzed as paired data, or as independent samples? Justify your answer.