Exploring break-points and interaction effects among predictors of the international digital divide

The deepening of the digital divide between countries has prompted international organizations and governments to work together toward reducing the problem over the next 15 years. However, such efforts will likely succeed only if they are based on a firm grasp of the divide's underlying causes. In this paper we report the results of a comprehensive analysis of the determinants of the international digital divide. Our results confirm many findings of past research, but also extend existing knowledge in important ways. By employing Multivariate Adaptive Regression Splines (MARS), we discover non-linearities and interaction effects among the predictors. We then articulate significant policy implications based upon these findings.