| Despite over two decades of research on media choice, influential theories (e.g., Daft & Lengel, 1986; Fulk, 1993) largely ignore the role of individual attitude in media choice. Likewise, relational maintenance researchers (Stafford, 2003) generally undertheorize the connection between content and medium in relational maintenance behavior. In light of these lacunae, this dissertation aims to achieve three goals. First, this project conceptualizes online communication attitude as a multidimensional cognitive/affective construct (Rokeach, 1968) that influences how people communicate online. The two studies reported here develop and validate a measure for assessing online communication attitude on eight dimensions. Second, confirmatory factor analysis tests a multitrait/multimethod decomposition model of relational maintenance behavior as measured by a modified version of the Canary and Stafford (1992) relational maintenance scale. This model effectively separates the content-specific and medium-specific components of manifest indicators of relational maintenance behavior. After establishing these measures and models, the final goal of this project tests a structural equation model specifying the relationship between online communication attitude, relational maintenance behavior, and friendship strength. Specifically, a two-group comparison examines differences in association between these constructs in both local and long-distance friendships. Results indicate that several online communication attitude dimensions predict media choice for relational maintenance behavior in local friendships, yet online communication attitude is a less potent predictor in long-distance friendships. Shared social networks and face-to-face maintenance predict friendship strength in both friendship groups, with positivity a predictor only in local friendships. Among the most important implications of this project are that individual attitude influences media choice in some contexts, and that medium-specific effects may account for some of the shared variance in relational maintenance behavior observed in previous confirmatory factor analyses on Canary and Stafford's relational maintenance scale. It is hoped that these results will contribute to refinement of relational maintenance theory and the development of theories that account for the multimodality of social life (Walther & Parks, 2002). |
