where α is the intercept, β1–β7 are regression coefficients of the independent variables and E represents the error term. Because of considerable correlations among the independent variables (see Table 4), high levels of multicollinearity were expected. The variance inflation factors (VIF) were calculated and found to be in the range of 1.5–3.5. In order to separate better the effects of the different independent variables a Ridge-regression was therefore conducted. In the case of multicollinearity, least squares estimates themselves are unbiased, but their variances are very large and possibly far from the true values. By adding a small bias (k) to the regression estimates, more reliable estimates can be obtained. A value of 0.6 for k was chosen, as this produced the best results. The regression results are reported in Table 5. From Table 5 it can be concluded that highly significant relationships exist between the two major user interface dimensions, navigation (Proposition 2) and design (Proposition 3), as well as the adapted SERVQUAL dimensions reliability (Proposition 4) and customization (Proposition 7), and customer responses.