Abstract
Prior studies examining humor in romantic relationships often assume that behavioral ratings obtained from different sources (i.e., self, partner, and observer) are interchangeable. However, it is unclear if humor ratings across sources reflect a common construct. The current study (N = 160 couples, M
age = 20 years, M
relationshiplength = 15 months) investigated whether partner and observer ratings of humor offer unique information beyond actor ratings. Additionally, the relative predictive validity of humor ratings was evaluated. In general, the results showed that ratings tend to converge across sources. In addition, partner and observer ratings generally did not explain unique variance beyond self‐ratings. These results suggest that rating sources for assessing humor are largely interchangeable, with a few exceptions. This study offers recommendations for the methodological consideration of rating sources regarding humor reports in close relationships.