Decoding University–Industry Collaboration: A SEM-ANN Quadruple Helix Approach

Abstract

University–industry collaboration (UIC) has received special emphasis from academicians and policymakers due to its potential for innovation diffusion and knowledge dissemination, leading to innovation ecosystem development and socio-economic advancement. Though extant literature has explored mechanisms to enhance university–industry collaboration, it has not investigated the quadruple helix model by integrating the role of academia, business firms, government, and civil society in fostering UIC. Grounded in the quadruple helix model, the objective of this research is to unveil the determinants of university–industry collaboration through developing an integrated framework. Data were gathered through a cross-sectional survey with 253 faculty members involved with the academia–industry collaboration research projects in Bangladeshi universities. To detect nonlinear relationships among variables, data were analyzed using a novel dual-staged structural equation modeling-artificial neural network (SEM-ANN) approach. The university’s innovation climate, mismatch of orientation in the academia–industry, and motivation-related constraints were found to have significant influence on university–industry collaboration (UIC). Besides, government support and input from civil society moderate the relationships between the predictors and UIC. However, the alignment of mutual goals does not have significant impact on harnessing UIC. Based on the normalized importance imputed from the ANN algorithm, the university’s innovation climate was proved to be the strongest predictor, followed by motivation-related constraint and mismatch of orientation between the university and industry. In light of the results, several insightful theoretical and practical implications are discussed for enhancing university–industry collaboration.