| Proliferation of theories and constructs that have attempted to explain information technology (IT) adoption have grown rapidly in the last decade. However, research is still needed in areas that have not been thoroughly investigated. For example, only a tiny fraction of the studies focused on IT adoption have been devoted to developing countries, which account for 85 percent of the worldwide population. An even smaller fraction of these studies has focused on small and medium sized enterprises (SMEs), which account for a high percentage of the economy generated in their respective countries. In addition, much of the research concerning SMEs and IT adoption, particularly those focused on developing countries, is descriptive in nature and lacks a strong theoretical basis.; In order to provide a solid, theory-based model that explains e-commerce adoption in an understudied segment of the population, this study examined the applicability of the Theory of Planned Behavior (TPB) and competing models to the prediction of the intention to adopt e-commerce among managers/owners of SMEs in Chile. To this end, data was collected on two occasions. A preliminary sample of the target population was randomly selected to elicit managers/owners' cognitions regarding the adoption of e-commerce. The results of this survey served as starting point to generate the theory-based instrument. After a pilot test, the instrument was sent to a larger sample of the same population to validate the applicability of the TPB, its extension, and competing models to give a more rigorous test of theory. Structural equation modeling was employed to test the hypotheses.; The results showed that the TPB explained 80 percent of the variance in behavioral intention. The Theory of Reasoned Action, a competing model of the TPB, proved to be as good as the TPB and could be even preferred for its parsimony. Attitude and subjective norm were found to be significant predictors of the intention to adopt e-commerce among Chilean managers. Perceived behavioral control, as a first-order factor, was not a significant predictor. However, when included as a second-order factor with controllability and self-efficacy as first-order factors, it weakly contributed to the prediction of intention. |
