ABSTRACT
The estimation results of the mixed logit model are reported in Tables 3-5.18 The coefficients for the respondent specific variables in Table 3 represent the effects relative to the SDA, e.g., the coefficient for the IP in Reykjavik (21.34∗) indicates that respondents in Reykjavik were less likely to vote for the IP than the SDA and that the difference is statistically significant at the 90% level. The effect for the LM is
156 I.H. Indridason
The Collapse: Economic Considerations in Vote Choice in Iceland 157
References
Appendix
The estimation results of the mixed logit model are reported in Tables 3-5.18 The coefficients for the respondent specific variables in Table 3 represent the effects relative to the SDA, e.g., the coefficient for the IP in Reykjavik (21.34∗) indicates that respondents in Reykjavik were less likely to vote for the IP than the SDA and that the difference is statistically significant at the 90% level. The effect for the LM is
156 I.H. Indridason
also statistically significant from the SDA but the table doesn’t show whether the effects for the IP and the LM are statistically different.
