Taking levels of analysis in humor more seriously: comment on Romero and Arendt

Humor is a phenomenon that can simultaneously coexist at the individual, dyadic, and group levels, making its measurement and conceptualization complex. In a recent field study, Romero and Arendt (2011) examined relationships between four humor styles (i.e., affiliative, self-enhancing, aggressive, self-defeating) and four outcomes (i.e., stress, satisfaction with co-workers, team cooperation, organizational commitment), however, the latter was apparently measured as a self-report at the individual level of analysis. Their interesting results indicated different humor styles can have either positive or negative effects on these outcome variables. However, if their operational definition--and hence their conceptualization--of humor is based on self-report by the initiator, it may be problematic to use it at the dyadic and group levels because it potentially mixes levels of analysis and may cause misalignment between data and theory. Cautions and implications for future research are discussed.