|
Abstract:
|
Research on relationships often do not refer to a single person but rather to two persons. Nonetheless, such data has been often analyzed by examining individuals in isolation, which falls short of capturing their truly interpersonal and nonindependent nature. Multi-level modeling, which accounts for nonindependence of responses on outcomes for the two members within the dyad and integrates the interrelationship between outcomes and predictors, was applied to augment the paired t-test and to extend the analysis of dyadic data from 73 males and their partners who participated in a clinical trial on erectile dysfunction. Among the results found were that patients and partners did not differ on treatment satisfaction (p-value = 0.13). The presentation highlights and illustrates methodology that is essential for theories about dyadic relationships to be tested adequately.
|
