Factors associated with implementation and usage of technology-based training: developing and testing a technology implementation model

The objectives of this dissertation were to (a) identify factors associated with implementation and use of technology-based training (TBT), (b) develop a model for TBT implementation, and (c) test the extent to which key portions of the model were supported by empirical data. Based on the literature, I developed a new model of TBT implementation, the Technology Implementation (TI) model. In order to test a portion of the model, I hypothesized that (a) TI model variables would account for a significant amount of variance in individual usage rates of the interactive video instruction (IVI) program at Federal Express, (b) several TI variables would not account for unique variance in IVI usage due to multicollinearity, and (c) the TI variables or a subset thereof would be significantly better predictors of IVI usage than variables from three other implementation models. Using multiple regression analysis, I found that the full set of TI variables did not account for a significant amount of variance in IVI usage (R$\sp2$ =.066, p =.086). Multicollinearity was found among the TI variables, but did not explain the nonsignificant full model regression findings. When subsets of independent variables were investigated, three TI variables (perceived relevance of the technology to the user's job, perceived net impact of using the technology, and perceived management support) and two Federal Express-specific variables (performance reviews and job knowledge test scores) accounted for a statistically significant amount of variance in IVI usage (R$\sp2$ =.092, p $<$.0001). Hierarchical regression F tests indicated that these five variables accounted for a statistically significant, though small, amount of variance in IVI usage beyond that accounted for by variables associated with mutual adaptation theory, the equity implementation model, and the technology acceptance model (even when test score and performance review were added to these models). While the research hypotheses were partially supported, the amount of variance in IVI usage accounted for was modest. Implications of these results are discussed, including potential confounds. Recommendations are provided for future research efforts and for corporate implementations of technology-based training.