Which candidates were the most likely to adopt and use Facebook in the 2006, 2008, and 2010 elections to the U.S. House of Representatives? This study affords an opportunity to examine the early adoption and dissemination of emerging technology tools in campaigns. To answers our research questions, we derive hypotheses primarily from the diffusion of innovations literature. Our multi-level analysis of over 2,400 candidates across three election cycles indicate that Facebook adoption diffused rapidly between 2006 and 2008 and continued to spread in 2010. Competition, money, and the level of education within the constituency were the most significant predictors of both adoption and more extensive implementation. Challengers were more likely to be early adopters, but incumbents used Facebook more extensively. Higher adoption rates by peers or competitors in the candidate’s own state and a propensity to adopt other campaign technology innovations are strong positive motivators for early adoption, but irrelevant to usage. These findings and our interviews with representatives from nearly 150 campaigns suggest that the medium has not changed the underlying campaign dynamic behind which candidates become early adopters and extensive users of new technologies nor has it changed significantly the importance of traditional campaign tools and tactics. While these findings are interesting in their own right, we make other important contributions to the future of study of online politics. We show that recent studies (whether in the US or other nations) have underestimated the influence of constituency level variables by estimating OLS regression models rather than a hierarchical linear or multi-level models; previous research has overlooked important variables such as peer adoption and past technology use to explain adoption; and that personal interviews with candidates and staff provide a richer understanding of the causal connection between candidates’ online behavior and the motives influencing that behavior.
