Table 3.
Two-Stage Least Squares (2SLS) Results
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Variables | A. second-stage regression (Dep.Var. = Case60) | |||||
| Individualism | 36.064*** | 40.294*** | 34.930*** | 39.408*** | 27.148** | 21.412 |
| (7.525) | (8.954) | (7.531) | (8.963) | (12.046) | (21.271) | |
| Dist_coast | 1.802 | 1.931 | 1.810 | 0.875 | ||
| (3.023) | (3.25) | (3.153) | (2.886) | |||
| Ruggedness | 157.506 | 137.37 | 102.075 | 200.140* | ||
| (98.488) | (106.041) | (105.159) | (102.347) | |||
| Precipitation | 0.247 | 0.24 | 0.213 | 0.375 | ||
| (0.192) | (0.193) | (0.185) | (0.266) | |||
| Age1564 | 29.213*** | 12.483 | 11.889 | − 41.220* | ||
| (10.696) | (16.706) | (15.657) | (21.270) | |||
| Density | − 0.071* | − 0.073 | − 0.071 | − 0.223** | ||
| (0.038) | (0.055) | (0.049) | (0.099) | |||
| Europe | 644.383* | 655.448** | ||||
| (381.561) | (310.121) | |||||
| Government Response | 13.327** | |||||
| (6.065) | ||||||
| GDP per capita | 0.037*** | |||||
| (0.011) | ||||||
| Quality of Government | − 1509.323 | |||||
| (945.464) | ||||||
| Personal Autonomy | − 27.408 | |||||
| (66.960) | ||||||
| N | 83 | 80 | 82 | 80 | 80 | 78 |
| R2 | 0.241 | 0.258 | 0.253 | 0.267 | 0.361 | 0.427 |
| B. First–stage estimates (Dep.Var. = Individualism) | ||||||
| Pronoun drop |
− 18.047*** (4.217) |
− 17.788*** (4.573) |
− 17.431*** (4.306) |
− 17.484 *** (4.785) |
− 17.314*** (4.620) |
− 11.084** (4.737) |
| Latitude |
0.736*** (0.094) |
0.682*** (0.121) |
0.772*** (0.096) |
0.720*** (0.126) |
0.763 (0.216) |
0.302 (0.187) |
| N | 83 | 80 | 82 | 80 | 80 | 78 |
| R2 | 0.616 | 0.626 | 0.625 | 0.631 | 0.631 | 0.733 |
| C. Diagnostic checks | ||||||
| 1st-stage F-statistic | 73.851 | 54.340 | 79.099 | 61.977 | 17.699 | 5.429 |
| 1st-stage partial R2 | 0.616 | 0.548 | 0.621 | 0.550 | 0.435 | 0.159 |
| Hansen J statistic (OID test) |
0.134 (p = 0.715) |
0.614 (p = 0.433) |
0.025 (p = 0.873) |
0.441 (p = 0.506) |
0.310 (p = 0.578) |
0.233 (p = 0.629) |
Robust standard errors are reported in parentheses. ***, ** and * indicate significance at the 0.01, 0.05, and 0.1 levels, respectively. Angola, Bangladesh, Bhutan, Cape Verde, Fiji, Honduras, Luxembourg, Malawi, Malta, Mozambique, Namibia, Nepal, Qatar, Senegal, Slovakia, Sri Lanka, Syria, and United Arab Emirates are excluded from the 2SLS regressions due to a lack of data on the pronoun drop variable (18 samples).
The missing values due to a lack of data on the pronoun drop are not missing at random since most of them are less developed countries and are in low IC. However, this removing these observations from the baseline specification still does not change the result (see Column (3) of Table 11 in Appendix for details).
Latvia and Morocco are further excluded in Column (6) due to the lack of data on Government Response