Table 2. Refinement using number of referenda on the same day.
| Identification: Number of referendaon the same day | Identification: Low turnout referendum | ||||
| (1) | (2) | (3) | (4) | (5) | |
| Years covered | 1848–2009 | 1848–2009 | 1848–2009 | 1884–2009 | 1848–2009 |
| Parliament suggests YES | 2.1695***(0.2763) | 2.2591***(0.2095) | 2.3493***(0.1986) | 2.7116***(0.1709) | 2.5194***(0.1566) |
| Parliament suggests YES * More than one referendum | 0.5883* (0.3458) | ||||
| More than one referendum | −0.4734(0.3164) | ||||
| Parliament suggests YES * More than two referenda | 0.7403** (0.3628) | ||||
| More than two referenda | −0.7052**(0.3250) | ||||
| Parliament suggests YES * More than three referenda | 0.8871** (0.4141) | ||||
| More than three referenda | −0.3194(0.3578) | ||||
| Parliament suggests YES * Low turnout referendum | 0.5391*** (0.1538) | 0.8113*** (0.1964) | |||
| Low turnout referendum | −0.1951(0.1260) | −0.2158(0.1582) | |||
| Number of referenda on the same day | 0.1740***(0.0229) | 0.0850**(0.0366) | |||
| Counterproposal | −0.2180(0.3051) | −0.2126(0.3056) | −0.1930(0.3080) | −0.7686***(0.0785) | −0.4511***(0.1032) |
| Turnout | −3.1623***(0.5265) | −3.9928***(0.8108) | |||
| Constituency fixed effects | YES | YES | YES | YES | YES |
| Decade fixed effects | YES | YES | YES | YES | YES |
| DE Parliament suggests YES | 49.1786 | 51.0306 | 46.4629 | 53.4686 | 53.0404 |
| DE Parliament suggests YES *More than one referendum | 9.3615 | ||||
| DE Parliament suggests YES *More than two referenda | 10.5967 | ||||
| DE (Parliament suggests YES) *(More than three referenda) | 15.7919 | ||||
| DE (Parliament suggests YES) *(Low turnout referendum) | 9.1936 | 14.0232 | |||
| Sample restriction | >2 referenda | >3 referenda | |||
| Clustering | Referendum | Referendum | Referendum | Constituency | Constituency |
| Pseudo R2 | 0.325 | 0.3272 | 0.3307 | 0.4539 | 0.494 |
| Brier | 0.1862 | 0.186 | 0.1851 | 0.1551 | 0.1459 |
| No. obs. | 14127 | 14127 | 14127 | 6329 | 4015 |
Notes: The dependent variable for all logit estimations is Constituency accepts referendum. Robust clustered standard error estimates for referenda (1–3) and constituencies (4–5) are reported throughout the table. DE = discrete effect in the predicted probability (see Table 1 and text for details). ***, **, and * indicate a mean significance level of below 1%, between 1 and 5%, and between 5 and 10%, respectively.