Weitere Kapitel dieses Buchs durch Wischen aufrufen
Based on Fountain’s technology enactment framework, the process model of computing change, and important concepts from environmental approaches to organizations, this chapter develops a sophisticated statistical model of the influence of organizational, institutional, and contextual factors on the functionality of government-wide websites as an instance of e-government success. Therefore, the key argument of this chapter is that organizational structures and processes, institutional arrangements, and environmental conditions have an impact on the selection, design, implementation and use of information technologies in government settings.
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten
Sie möchten Zugang zu diesem Inhalt erhalten? Dann informieren Sie sich jetzt über unsere Produkte:
PLS analysis was performed using PLS-Graph version 3.00.
Partial least squares (PLS) is a structural equation modeling (SEM) statistical approach for modeling complex multivariable relationships among observed and latent variables (Esposito, Trinchera and Amato 2010). SEM has become the ultimate strategy in validating instruments and testing linkages between constructs (Henseler, Ringle and Sinkovics 2009). SEM attempts to understand the causal relationships between theoretical constructs and how well each construct is captured by its indicators or manifest variables. As a SEM technique, PLS can simultaneously test the structural model (relationships between constructs) and the measurement model (relationships between indicators and their corresponding constructs) (Barclay et al. 1995; Gil-Garcia 2008; Hulland 1999; Tenenhaus, Esposito, Chatelin and Lauro 2005). The PLS technique was considered an appropriate approach due to the nature of the complex relationships among the variables (Gil-Garcia 2005a). For a detailed description of the research design and methods, see Appendix A at the end of the book.
In addition, there are three reasons to keep this indicator. First, Chin (1998) mentions that loadings as low as 0.5 should be kept as long as there are other good indicators in the scale. In fact, other authors suggest a cut-off of 0.5 (Hair, Anderson, Tatham, and Black 1998) or 0.6 (SAS 1990) for loadings. Second, PLS estimates are consistent at large (Wold 1982). According to this principle “the estimates will approach the ‘true’ latent variable scores as both the number of indicators per block and the sample size increase.” (Chin and Newsted 1999, p. 329). Third, this indicator makes both theoretical and practical sense.
Following the logic of dummy variables, the reference category for these three possible outcomes (entirely in-house development, entirely outsourcing, and combination of both) was a combination of both. Therefore, the negative statistically significant coefficients of “entirely in-house development” and “entirely outsourcing” indicates that these two strategies have a negative impact in comparison to the reference category (combination of both).
- Identifying Electronic Government Success Factors: A Statistical Analysis
J. Ramon Gil-Garcia
- Springer US
- Chapter 3
ec4u, Neuer Inhalt/© ITandMEDIA