likely to adopt ABC.23 All of the significant variables listed above are in the direction predicted. The
Chi-square statistic shown in Table 7 is comparable to the overall F-test in multiple regression. The
model is statistically significant at the .000 level. The Hosmer and Lemeshow goodness of fit value (.574)
measures the correspondence of the actual and predicted values of the dependent variable. This statistic
tests the hypothesis that the observed data are significantly different from the predicted values. Thus, a
non-significant statistic indicates that the model does not differ significantly from the observed data (Hair
et al., 1998).24 Nagelkerke R2 (.515) attempts to quantify the proportion of explained “variation” in the
logistic regression model. It is similar in intent to the R2 in a linear regression model (Norusis, 2000).