• DocumentCode
    115377
  • Title

    Structural Equation Modelling based data fusion for technology forecasting: A National Research and Education Network example

  • Author

    Staphorst, Leon ; Pretorius, Leon ; Pretorius, Tinus W.

  • Author_Institution
    Council for Sci. & Ind. Res., Brummeria, South Africa
  • fYear
    2014
  • fDate
    27-31 July 2014
  • Firstpage
    2908
  • Lastpage
    2917
  • Abstract
    This paper presents an example model instantiation of Staphorst, Pretorius and Pretorius´ framework for Structural Equation Modelling (SEM) based Data Fusion (DF) for Technology Forecasting (TF) in the National Research and Education Network (NREN) technology domain. The paper´s example NREN model instantiation is constructed through deductive reasoning from knowledge gained during action research in the South African National Research Network (SANReN), as well as secondary data from TERENA´s NREN compendiums for global NREN infrastructure and services trends. A variety of technology related measurements are employed in the example NREN model instantiation as indicators for technology related model constructs, such as the level of core network traffic in an NREN. Indicators for context related model constructs include, amongst others, the range of institutions an NREN is mandated to connect. For confirmatory purposes the secondary data published by TERENA in its yearly NREN compendium series is then used in the Partial Least Squares (PLS) regression analysis to determine the indicator loadings and path coefficients of the example NREN model instantiation. A reliability and validity analysis of the example NREN model instantiation is also considered.
  • Keywords
    deductive databases; least squares approximations; production engineering computing; regression analysis; sensor fusion; technological forecasting; technology management; NREN technology; PLS regression analysis; South African national research network; TERENA NREN compendiums; data fusion; deductive reasoning; national research and education network technology; partial least squares regression analysis; structural equation modelling; technology forecasting; Context; Data models; Government; Load modeling; Loading; Mathematical model; Measurement;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Management of Engineering & Technology (PICMET), 2014 Portland International Conference on
  • Conference_Location
    Kanazawa
  • Type

    conf

  • Filename
    6921138